International Journal of Retail & Distribution Management

Kim, J.-H. and Kim, M. (2016), “Luxury fashion retailers and consumers’ perceptions of luxury fashion websites”, Proceedings of Academic of Marketing Science (AMS) World Marketing Congress, Paris, July. Kim, J.-H. and Zhang, B. (2015), “Attitude and purchase intent for luxury fashion goods: cultural differences between Americans and Chinese”, International Journal of Costume and Fashion, Vol. 15 No. 1, pp. 19-37. Kim, J.-H., Kim, M. and Kandampully, J. (2011), “The impact of e-retail environment characteristics on esatisfaction and purchase intent”, International Journal of Service Science, Management, Engineering, and Technology, Vol. 2 No. 3, pp. 1-19. Kluge, P.N. and Fassnacht, M. (2015), “Selling luxury goods online: effect of online accessibility and price display”, International Journal of Retailing & Distribution Management, Vol. 43 Nos 10/11, pp. 1065-1082. Ko, E.K., Costello, J.P. and Taylor, C.R. (2017), “What is a luxury brand? A new definition and review of the literature”, Journal of Business Research, available at: www.sciencedirect.com/science/article/ pii/S0148296317302850?via%3Dihub Korgaonkar, P., Petrescu, M. and Becerra, E. (2014), “Shopping orientations and patronage preferences for internet auctions”, International Journal of Retail & Distribution Management, Vol. 42 No. 5, pp. 352-368. Lee, G.-G. and Lin, H.-F. (2005), “Customer perceptions of e-service quality in online shopping”, International Journal of Retail & Distribution Management, Vol. 33 No. 2, pp. 161-176. Levy, M. and Weitz, B.A. (2001), Retailing Management, 4th ed., McGraw-Hill/Irwin, Boston, MA. Lissitsa, S. and Kol, O. (2016), “Generation X vs. generation Y – a decade of online shopping”, Journal of Retailing and Consumer Services, Vol. 31, pp. 304-312. Liu, X., Burns, A.C. and Hou, Y. (2013), “Comparing online and in-store shopping behaviour towards luxury goods”, International Journal of Retailing & Distribution Management, Vol. 41 Nos 11/12, pp. 885-900. Loiacono, E., Watson, R. and Goodhue, D. (2002), “WEBQUAL: a measure of website quality”, in Evans, K. and Scheer, L. (Eds), Marketing Educators’ Conference: Marketing Theory and Applications, Vol. 13, pp. 432-437. McKinsey and Company (2015), “The opportunity in online luxury fashion”, available at: www. mckinsey.com/business-functions/marketing-and-sales/our-insights/the-opportunity-in-onlineluxury-fashion (accessed May 4, 2017). Madu, C.N. and Madu, A.A. (2002), “Dimensions of e-quality”, International Journal of Quality & Reliability Management, Vol. 19, pp. 246-258. Mummalaneni, V., Meng, J. and Elliott, K.M. (2016), “Consumer technology readiness and e-service quality in e-tailing: what is the impact of predicting online purchasing?”, Journal of Internet Commerce, Vol. 15 No. 4, pp. 311-331. Noble, S.M., Haytko, D.L. and Phillips, J. (2009), “What drives college-age Generation Y consumers?”, Journal of Business Research, Vol. 62 No. 6, pp. 617-628. Nueno, J.L. and Quelch, J.A. (1998), “The mass marketing of luxury”, Business Horizons, Vol. 41, No. 6, pp. 61-68. Nunnally, J.C. and Bernstein, I.H. (1994), Psychometric Theory, McGraw-Hall, New York, NY. Nwankwo, S., Hamelin, N. and Khaled, M. (2014), “Consumer values, motivation and purchase intention for luxury goods”, Journal of Retailing and Consumer Services, Vol. 21 No. 5, pp. 735-744. Okonkwo, U. (2007), Luxury Fashion Branding, Palgrave Macmillan, Basingstoke. Okonkwo, U. (2010), Luxury Online: Styles, Systems, Strategies, Palgrave Macmillan, Basingstoke. Parasuraman, A., Zeithaml, V.A. and Malhotra, A. (2005), “E-S-QUAL: a multiple-item scale for assessing electronic service quality”, Journal of Service Research, Vol. 7 No. 3, pp. 213-233. 238 IJRDM 47,2

Phau, I. and Prendergast, G. (2000), “Consuming luxury brands: the relevance of the rarity principle”, Journal of Brand Management, Vol. 8 No. 2, pp. 122-138. Piercy, N. (2014), “Online service quality: content and process of analysis”, Journal of Marketing Management, Vol. 30 Nos 7-8, pp. 747-785. Prateek, K. (2017), “Service quality scales in online retail: methodological issues”, International Journal of Operations & Production Management, Vol. 37 No. 5, pp. 630-663. Punj, G. (2012), “Income effects on relative importance of two online purchase goals: Saving time versus saving money?”, Journal of Business Research, Vol. 65 No. 5, pp. 634-640. Rolland, S. and Freeman, I. (2010), “A new measure of e-service quality in France”, International Journal of Retail & Distribution Management, Vol. 38 No. 7, pp. 497-517. Roper, S., Caruana, R., Medway, D. and Murphy, P. (2013), “Constructing luxury brands: exploring the role of consumer discourse”, European Journal of Marketing, Vol. 47 Nos 3/4, pp. 375-400. Rosen, D.E. and Purinton, E. (2004), “Website design: viewing the web as a cognitive landscape”, Journal of Business Research, Vol. 57 No. 7, pp. 787-794. Ruane, L. and Wallace, E. (2013), “Generation Y female online: Insights from brand narratives”, Qualitative Market Research: An International Journal, Vol. 16 No. 3, pp. 315-335. Sabiote, C.M., FrÍas, D.M. and Castañeda, J.A. (2012), “E-service quality as antecedent to e-satisfaction: the moderating effect of culture”, Online Information Review, Vol. 36 No. 2, pp. 157-174. Soh, C.Q.Y., Rezaei, S. and Gu, M.-L. (2017), “A structural model of the antecedents and consequences of generation Y luxury fashion goods purchase decisions”, Young Consumers, Vol. 18 No. 2, pp. 180-2-4. Srinivasan, S.S., Rolph, A. and Kishore, P. (2002), “Customer loyalty in e-commerce: an exploration of its antecedents and consequences”, Journal of Retailing, Vol. 78 No. 1, pp. 41-50. Statista (2018), “Total sales of the luxury goods market worldwide from 2016 to 2017, by product category (in billion euros)”, Statista, available at: www.statista.com/statistics/245655/total-salesof-the-luxury-goods-market-worldwide-by-product-category/ (accessed September 29, 2018). Swaid, S.I. and Wigand, R.T. (2009), “Measuring the quality of e-service: scale development and initial validation”, Journal of Electronic Commerce Research, Vol. 10 No. 1, pp. 13-28. Tűrk, B., Scholz, M. and Berresheim, P. (2012), “Measuring service quality in online luxury goods retailing”, Journal of Electronic Commerce Research, Vol. 13 No. 1, pp. 88-103. Tynan, C., McKechnie, S. and Chhuon, C. (2010), “Co-creating value for luxury brands”, Journal of Business Research, Vol. 63 No. 11, pp. 1156-1163. Valaei, N. and Nikhashemi, S.R. (2017), “Generation Y consumers’ buying behaviour in fashion apparel industry: a moderation analysis”, Journal of Fashion Marketing and Management: An International Journal, Vol. 21 No. 4, pp. 523-543. Vigneron, F. and Johnson, L.W. (2004), “Measuring perceptions of brand luxury”, Journal of Brand Management, Vol. 11 No. 6, pp. 484-506. Wiedmann, K.P., Hennigs, N. and Siebels, A. (2009), “Value-based segmentation of luxury consumption behaviour”, Psychology & Marketing, Vol. 26 No. 7, pp. 625-651. Wolfinbarger, M. and Gilly, M.C. (2003), “ETailQ: dimensionalizing, measuring and predicting etail quality”, Journal of Retailing, Vol. 79, pp. 183-198. Yang, F.X. and Lau, V.M.C. (2015), “LuXurY hotel loyalty-A comparison of Chinese Gen X and Y tourists to Macau”, International Journal of Contemporary Hospitality Management, Vol. 27 No. 7, pp. 1685-1706. Yang, W. and Mattila, A.S. (2014), “Do affluent customers care when luxury brands go mass? The role of product type and status seeking on luxury brand attitude”, International Journal of Contemporary Hospitality Management, Vol. 26 No. 4, pp. 526-543. 239 Imperative challenge for luxury brands

Yang, Z., Jun, M. and Peterson, R.T. (2004), “Measuring customer perceived online service quality: Scale development and managerial implications”, International Journal of Operations & Production Management, Vol. 24 No. 11, pp. 1149-1174. Yarimoglu, E.K. (2017), “Demographic differences on service quality and perceived value in private online shopping clubs”, Journal of Strategic Marketing, Vol. 25 No. 3, pp. 240-257. Yoo, B. and Donthu, N. (2001), “Developing a scale to measure the perceived quality of an Internet shopping site (SITEQUAL)”, Quality Journal of Electronic Commerce, Vol. 2 No. 1, pp. 31-46. Zehir, C. and Narcikara, E. (2016), “E-service quality and e-recovery service quality: effects on value perceptions and loyalty intentions”, Procedia-Social and Behavioral Sciences, Vol. 229, pp. 427-443. Zeithaml, V.A., Parasuraman, A. and Malhotra, A. (2000), “A conceptual framework for understanding e-service quality: implications for future research and managerial practice”, Working Paper No. 00-115, Marketing Science Institute, Cambridge, MA. Zeithaml, V.A., Parasuraman, A. and Malhotra, A. (2002), “Service quality delivery through web sites: a critical review of extant knowledge”, Journal of the Academy of Marketing Science, Vol. 30 No. 4, pp. 358-371. Zhang, B. and Kim, J.-H. (2013), “Luxury fashion goods and Chinese consumer: factors affecting attitude and purchase intent”, Journal of Retailing and Consumer Services, Vol. 20 No. 1, pp. 68-79. Zhang, M., Huang, L., He, Z. and Wang, A.G. (2015), “E-service quality perceptions: an empirical analysis of the Chinese e-retailing industry”, Total Quality Management, Vol. 26 No. 12, pp. 1357-1372. Zhang, X. and Prybutok, R. (2005), “A consumer perspective of e-service quality”, IEEE Transactions on Engineering Management, Vol. 52 No. 4, pp. 461-477. Further reading Han, Y.J., Nunes, J.C. and Dreze, X. (2010), “Signaling status with luxury goods: the role of brand prominence”, Journal of Marketing, Vol. 74 No. 4, pp. 15-30. Kluge, P.N., Königsfeld, J.A., Fassnacht, M. and Mitschke, F. (2013), “Luxury web atmospherics: an examination of homepage design”, International Journal of Retail & Distribution Management, Vol. 41 Nos 11/12, pp. 901-916. Milnes, H. (2015), “How 3 high-end brands balance luxury with e-commerce”, available at: https:// digiday.com/marketing/3-high-end-brands-balance-luxury-e-commerce/ (accessed May 6, 2017). Ojasalo, J. (2010), “E-service quality: a conceptual model”, International Journal of Arts and Sciences, Vol. 3 No. 7, pp. 127-143. Scott, M. (2015), “Luxury brands and the social campaign”, The New York Times, available at: www.nytimes.com/2015/12/02/fashion/luxury-brands-and-the-social-campaign.html?_r=0 (accessed May 23, 2017). Vigneron, F. and Johnson, L.W. (1999), “A review and a conceptual framework of prestige-seeking consumer behavior”, Academy of Marketing Science Review, Vol. 1, pp. 1-15. Walker, R. (2016), “25 predictions for the luxury goods industry in 2016”, available at: http://blog. euromonitor.com/2016/01/25-predictions-luxury-goods-industry-2016.html Walters, J. (2016), “The conflict between digital democratization and luxury brands”, Retail Customer Experience.com”, available at: www.retailcustomerexperience.com/articles/conflict-betweendigital-democratization-and-luxury-brands/ (accessed May 10, 2017). Wiedmann, K.P., Hennigs, N. and Siebels, A. (2007), “Measuring consumers’ luxury value perception”, Academy of Marketing Science Review, Vol. 7 No. 7, pp. 1-21. 240 IJRDM 47,2

Appendix. Survey Questionnaire utilized for the study 241 Imperative challenge for luxury brands

242 IJRDM 47,2

243 Imperative challenge for luxury brands

Corresponding author Jung-Hwan Kim can be contacted at: [email protected] For instructions on how to order reprints of this article, please visit our website: www.emeraldgrouppublishing.com/licensing/reprints.htm Or contact us for further details: [email protected] 244 IJRDM 47,2

Direction of luxury fashion retailers’ post-entry expansion – the evidence from China Huifeng Bai Liverpool Business School, Liverpool John Moores University, Liverpool, UK Julie McColl Glasgow Caledonian University, Glasgow, UK Christopher Moore New College Lanarkshire, Motherwell, UK Weijing He University of Jinan, Jinan, China, and Jin Shi Saitama University, Saitama, Japan Abstract Purpose – This empirical study, from the international retailing perspective, examines the direction of retailers’ further expansion after initial entry into overseas host market in the context of the luxury fashion retail market in China. Design/methodology/approach – The research adopts qualitative multiple case studies. Findings – After initial entry into China, luxury fashion retailers further expand their retail operations through three directional patterns: cautious, regional and countrywide expansions. The stepwise expansion from tier-1 to tier-2 and tier-3 cities remains popular; however, the importance of the tier system of Chinese cities has been weakened because tier-3 cities in affluent regions are perceived to have more potential than some tier-2 cities in less developed regions. The retailers assess a potential local market through interrelated criteria, including location and strategic importance, economic development, available store locations and staff, a high degree of urbanisation and tourism, debatable favourable policies and offers, and popularity of e- and mcommerce. There is a positive relationship between popularity of e- and m-commerce in a city and the potential of that city to run brick-and-mortar stores. Originality/value – The paper offers an insight into the current international retailing literature by examining the direction of luxury fashion retailers’ further expansion after their initial market entry. Particularly, the research considers a set of criteria which can be used to assess a potential local market, and the impact of e- and m-commerce on local market choices for brick-and-mortar stores. Keywords Retailer internationalisation, Post-entry expansion, Direction, Luxury fashion, Omni-channel retail, Market development Paper type Research paper 1. Introduction International retailing has become an increasingly important and popular research field since the 1980s, because of the transfer of market power from manufacturers and wholesalers to retailers, and the acceleration of retailer internationalisation activities (Swoboda et al., 2009). In particular, fashion and luxury fashion retailers, whose business activities include manufacturing, distribution, communication and service provision, have achieved considerable worldwide success since the early 1990s (Hennigs et al., 2015). Recently, the landscape of the international luxury fashion market had been influenced by rapidly growing continent-sized emerging markets, in particular, China (Ye et al., 2018). In the past two decades, Luxury fashion retailers’ postentry expansion 223 The current issue and full text archive of this journal is available on Emerald Insight at: https://www.emerald.com/insight/0959-0552.htm Received 19 June 2020 Revised 26 June 2020 21 September 2020 Accepted 24 October 2020 International Journal of Retail & Distribution Management Vol. 49 No. 2, 2021 pp. 223-241 © Emerald Publishing Limited 0959-0552 DOI 10.1108/IJRDM-06-2020-0218

luxury fashion retailers in China have not only strengthened their retail operations in tier-1 national capitals, but they have also extended their retail operations into wider ranges of tier-2 regional and/or provincial capitals and tier-3 non-capital cities as well as e-commerce (Baiet al., 2017). Indeed, the value of the international luxury fashion market reached V281 billion in 2019, of which over one-third was contributed by the Chinese market (Bain & Co, 2019). Meanwhile, in the international retailing literature, recent studies have focused on more specific issues, such as branding strategies (Christmann et al., 2016), the internationalisation of small retailers (Hutchinson and Quinn, 2011), international expansion into emerging markets (Gomes et al., 2018), and omni-channel retailing (Ye et al., 2018). The classic sixquestion research agenda for retailer internationalisation (what, who, why, where, how and when), and in particular the direction of retailer internationalisation (where), however, still remain important, because the choice of host markets helps to understand retailers’ internationalisation strategies, that is, perceived psychic distance, resource commitment and external trading conditions (Schu and Morschett, 2017). Nevertheless, the majority of the previous literature has focused on the country-level market, rather than post-entry expansion strategies and activities after initial market entry (Frasquet et al., 2013). Indeed, retailer internationalisation is a dynamic and is a continuous process, which is not only about foreign market selection and entry but is also about post-entry further expansion and ongoing management and operations (Bai et al., 2017). Therefore, this empirical study aims to examine the direction of luxury fashion retailers’ post-entry expansion direction after their initial entry into a host market in China. The following section reviews the literature in relation to the definition of luxury fashion retailers and their post-entry expansion direction. 2. Literature review 2.1 Defining luxury fashion retailers The democratisation of luxury in modern society has seen increasing numbers of marketers adopt the philosophy of luxury into their branding strategies for both tangible goods and intangible services in order to fuel demand and persuade target consumers to trade up, communicating exclusivity and adding value for example through material sourcing, production processes, distribution and communication (Kapferer and Valette-Florence, 2016). The growth of luxury market potential has resulted in expansion into international markets and increased global success and has led academics to focus their research on luxury brands, and particularly luxury brand management (Kapferer, 2015), brand extension (Boisvert and Ashill, 2018), and consumer behaviour (Shukla et al., 2016), especially online shopping activities (Liu et al., 2013), and counterfeit consumption (Chen et al., 2015). Despite this level of interest in the luxury goods and market, a universally accepted definition for luxury brand has yet to be established because of a high degree of subjectivity and the strong influence of the sociocultural context, various expectations about the quality of luxury goods and services, and different levels of premium prices within different countries or cultural zones (Laforet and Chen, 2012). Most recently, Ko et al. (2019) evaluated several high impact studies relating to the luxury market and re-defined the luxury brands from the consumer perspective as: “a branded product or service that consumers perceive to: (1) be high quality; (2) offer authentic value via desired benefits, whether functional or emotional; (3) have a prestigious image within the market built on qualities such as artisanship, craftsmanship, or service quality; (4) be worthy of commanding a premium price; and (5) be capable of inspiring a deep connection, or resonance, with the consumer”. (p.406) IJRDM 49,2 224

Although the definition re-emphasised the majority of criteria of luxury brands in general, some elements can be challenged, particularly in the context of fashion retailing. The parameters that qualify a brand to be luxury brand are dynamic in nature. The exclusivity of luxury brand image can be diluted by easy accessibility and over distribution (Kapferer, 2015), and the inconsistency between a brand’s country-of-origin and place-ofproduction (Arora et al., 2015). It is also difficult for a luxury brand to maintain its luxury status in the face of fast-moving markets and increasingly sophisticated consumers, witnessed by “anti-logo” consumption patterns, which reveals that luxury brands with conspicuous logos are decreasing in popularity (Makkar and Yap, 2018). Moreover, overextension, especially downward extension of luxury brands and the upward extension of premium brands, has blurred the boundary between luxury and premium (Spiggle et al., 2012). Because fashion retailing is one of the most important and popular business sectors which widely employs luxury branding strategies, Moore et al. (2010) suggested to understand luxury brands from a more practical perspective. They define luxury fashion retailers as those who: “distribute clothing, accessories and other lifestyle products which are: (1) exclusively designed and/or manufactured by or for the retailer; (2) exclusively branded with a recognised insignia, design handwriting or some other identifying device; (3) perceived to be of a superior design, quality and craftsmanship; (4) priced significantly higher than the market norm; (5) sold within prestigious retail setting”. (p.143) Since the scope of this study is international retailing, this definition is considered to be the most appropriate due to detailed operational dimensions employed by luxury fashion retailers. This paper focuses on luxury fashion (manufacturing) retailers such as Louis Vuitton (Guercini and Milanesi, 2017), rather than high-end fashion (distributing) retailers such as Harrods or premium fashion (manufacturing) retailers such as COACH. 2.2 Luxury fashion retailer post-entry expansion direction Similar to other categories of retailer, the international expansion of luxury fashion retailers is a geographical process (Siebers, 2017). Therefore, their evaluation and decision-making about host markets provides important clues as to the challenges that they expect to face and how may find solutions to these challenges (Ojala, 2015). Luxury fashion retailers assess target markets through either the one-stage approach which mainly involves screening macroeconomic indices or by multi-stage models which are stepped decision based on market evaluation (Frasquet et al., 2018). Within the classic international business literature, psychic distance is regarded as the core of the incremental internationalisation approach, which explores the challenges of foreign markets, including culture, language, education, business practice, political and legal systems, economic development and industry structure (Evans et al., 2008). The Uppsala model identifies that companies initially expand into markets psychically proximate with their home markets, and then extend into psychically distant markets, increasing their commitment to the host markets when they become more confident having gained local knowledge and international experience (Vahlne and Johanson, 2017). Indeed, in order to minimise the risks associated with internationalisation and maximise control over operations abroad, generic retailers even fast fashion retailers are keen to adopt a “stepwise” expansion Luxury fashion retailers’ postentry expansion 225

strategy to internationalise, from geographically/culturally proximate host markets to distant markets, for example ZARA, GAP and H&M (Lopez and Fan, 2009). Nevertheless, the importance of psychic distance in modern international markets has been weakened due to integration of global culture and economics, and improved communication technology (Schu and Morschett, 2017). Not all retailers, however, adopt an incremental path to internationalisation, particularly luxury fashion retailers (Chevalier and Gutsatz, 2020). Increasingly homogeneous global customer preferences, distinctive products and the global attractiveness of brands can each prompt worldwide expansion of luxury fashion retailers, who are less inhibited by issues of cultural and geographical accessibility and instead seek to maximise success through aggressive globalisation strategies which involve aspects of product desirability, lifestyle, image and niche opportunities, such as the “New York, London, and Paris syndrome” (Moore et al., 2010). Despite contribution, most of prior studies focused on developed markets rather than emerging markets (Yu and Ramanathan, 2012). They also focused on initial foreign market entry rather than explaining the criteria used to assess and evaluate local markets after initial entry (Bai et al., 2017). According to a small body of literature, which focuses on grocery retailers in China, it is identified that multinational supermarket chains developed their business operations in China through a countrywide expansion pattern, which simply followed the tier-system of Chinese cities (Wang and Li, 2012). Indeed, Wang and Liu (2008) argue that foreign supermarket chains extended their operations to tier-2 and then tier-3 cities from tier-1 national capital cities or entry city. Siebers (2017) confirmed such post-entry expansion patterns, and highlighted the regional differences in terms of socioeconomic factors within China. These valuable studies identified the regional differences and provided a relatively comprehensive understanding of foreign retailers’ post-entry expansion direction in China. Unfortunately, they are limited in explaining the activities of luxury fashion retailers due to different branding strategies and operations (Kapferer, 2015). Moreover, Chevalier and Lu (2009) identified that luxury fashion retailers typically enter into the US market from East and West costal metropolis, and then simply expand further into middle regions. However, the regional differences in China are far more complicated than in the US market (Liu et al., 2016). In the context of the luxury fashion retail market in the UK, Moore et al. (2000) developed a four-step business model employed by internationalising luxury fashion retailers entering and developing business within the UK (Figure 1). Luxury fashion retailers initially entered into UK through wholesaling agreement with established and high-end department stores in London, because this low-cost and low-risk entry mode helps them to generate cash flow, customer loyalty and market intelligence. Once created demands within the capital cities, they would expand to provincial capitals through the department stores’ networks. Having gained success through wholesaling agreements, they enhanced engagement with capital cities by opening couture/Ready-To-Wear (RTW) flagship stores located in the premium shopping districts. Despite the extremely high cost, such flagship stores within capital cities have not only help them to generate profits through relatively more standardised RTW, but more importantly to help them to raise brand awareness and even loyalty. They then extended further by opening diffusion brand flagship stores in capital cities, which maximised the availability of diffusion. Finally, they developed diffusion stores in key provincial cities, such as Manchester and Glasgow. The study offers depth insight into the where and how of luxury fashion retailers entering a host market, and the further development of operations within the country. However, it was developed within the UK, where high-end department stores are popular and regarded as an appropriate shopping destination for luxury fashion brands. This kind of market infrastructure offers foreign luxury fashion retailers opportunities to adopt low-cost and low-risk entry modes and to further expand via an incremental process while the market infrastructures in emerging markets are largely different. Bai et al. (2017) identified that IJRDM 49,2 226

luxury fashion retailers are keen to adopt high-cost modes to enter into China because of the lack of high-end department stores and the desire to fully control their brands and business practices. In terms of store formats, stand-alone stores located in luxury malls and high-end hotels are more suitable for foreign luxury fashion retailers in China, because of the lack of high-end department stores and premium shopping districts (Lu, 2012). Post-entry expansion beyond provincial cities had not been considered by this study. In terms of geographic size and domestic regional differences relating to economies, culture, urbanisation and market infrastructure, the UK market is not as complicated as the Chinese market (Bai et al., 2018). Therefore, the first research question is: R1. Where and how have internationalising luxury fashion retailers expanded after their initial market entry into China? Moore et al. (2000) only identified the post-entry expansion from capital cities to key provincial cities and did not offer the criteria employed by luxury fashion retailers to assess and select local markets for further expansion. Therefore, the second research question is: R2. What are the criteria employed by internationalising luxury fashion retailers to assess local market for their physical stores? Finally, the impact of e-commerce on luxury fashion retailers’ operation and brick-and-mortar store portfolio was not considered, probably because the impact and popularity of e-commerce at that stage was not well developed. The rapid development of e-commerce has influenced luxury fashion retailers’ internationalisation strategies and offered effective opportunities for them to explore and to build an insightful understanding of local markets in advance of actual market entry (Bai, 2018). Therefore, the third research question is: Stage 1 Stage 2 Stage 3 Stage 4 Launch couture/RTW via wholesale in capital city department stores Open couture/RTW flagships in capital cities Open diffusion brand flagship stores in capital Open diffusion stores in key provincial cities (leading to …) (leading to …) Extend wholesale availability of RTW to key stores in other major cities Maximum availability of diffusion brands via wholesale within capital & provincial cities Figure 1. The four stages of luxury fashion retailers market development in the UK Luxury fashion retailers’ postentry expansion 227

R3. How has e-commerce influenced foreign luxury fashion retailers’ decisions about local markets for brick-and-mortar stores in China? The next section will justify the research methodology and discuss the research design according to the above three research questions. 3. Research methodology Since this study’s research topic remains under researched, it is necessary to obtain a relatively comprehensive understanding of luxury fashion retailers’ post-entry expansion direction in China. Therefore, the researchers regarded an interpretivist qualitative multiple case study as the most appropriate methodology, focusing on “what”, “how”, and “why” research questions (Creswell and Creswell, 2018). It is also capable of analysing data within each case and/or across cases, and gaining an understanding about the similarities and differences between cases, therefore enhancing the validity and reliability of the findings (Yin, 2018). Indeed, multiple case studies have been widely employed in the luxury fashion marketing research field (Liu et al., 2016). In order to ensure sufficiency of strategic and operational experience, the target sample for this research was evaluated and selected by means of three criteria: originating from foreign countries; having operations in at least two cities in China; and having at least five years of experience. The database then was developed from three major sources. The lists of members of reputable luxury committees were employed, including Comit
e Colbert (France), Fondazione Altagamma (Italy), Walpole and British Council of Fashion (UK) and the Council of Fashion Designers of America (US). The directories of the top ten luxury malls in mainland China and the Peninsula Hotel Beijing were employed. The marketing reports published by established organisations were evaluated, including Bain & Co (2017), Deloitte (2017), and McKinsey & Co (2017). In the end, 76 target luxury fashion retailers were identified. The 76 target retailers were initially contacted by the cover letters, which were sent by e-mail, as well as mailed to the head office in their home countries and the regional head office in Hong Kong, Beijing, or Shanghai in May 2018. By the end of September 2018, 14 retailers agreed to participate. Since the researchers had identified that the samples had reached saturation, 30 executive interviews in total took place between October 2018 and July 2019, lasting between 45 and 70 minutes. All participating retailers and the interviewees were coded in order to fulfil the confidential agreements and were alternatively described through country of origin, retail formats, ownership structures and interviewees’ job titles. Moreover, all interviewees listed in Table 1 were capable of providing the necessary information on decision making about policy making and strategic implementation. In order to minimise researchers’ and interviewees’ subjective bias and enhance the validity of the data, the researchers utilised secondary data including annual reports, internal documents and marketing reports. Subsequently, the qualitative data was analysed using Thematic Analysis in NVIVO 10, which allowed for the development of codes and references that were retained in an index system so that the data could be carefully identified and explored. The findings will be presented and discussed in the next section. 4. Research findings 4.1 Luxury fashion retailers’ post-entry expansion direction in China This study identified that post-entry expansion patterns of the participant luxury fashion retailers are different (Table 2). Figure 2 illustrates the three major patterns of participant luxury fashion retailers’ postentry expansion in China. The first is cautious expansion, through which the retailers IJRDM 49,2 228

Country of origin Retail format Ownership Year and city of China’s entry Entry mode E-commerce Number of interviews and interviewee position A France Design house Private 1996 and Beijing Organic growth Yes 2 and Managing Director, Retail Operations Director B France Design house Subsidiary 1992 and Beijing Organic growth Yes 2 and Vice President, Retail Director C Germany Accessories and leather goods Subsidiary 1996 and Beijing Organic growth Yes 2 and Managing Director, Wholesale Manager D Germany Design house Private 1994 and Shanghai Franchising Yes 2 and Business Development Director, Retail Director E Italy Design house Subsidiary 2007 and Shanghai Organic growth Yes 2 and Vice President, Brand Manager F Italy Design house Private 1995 and Beijing Franchising Yes 2 and Managing Director, Wholesale Manager G Italy Design house Subsidiary 1993 and Beijing Franchising Yes 3 and Retail Manager, Managing Director, Brand Manager H Spain Design house Subsidiary 2001 and Shanghai Organic growth Yes 2 and Vice President, Managing Director I Switzerland Design house Subsidiary 1991 and Beijing Organic growth Yes 3 and Vice President, Digital Marketing Director, Retail Director J Switzerland Jewellery and watches Private 1999 and Beijing Organic growth No 2 and Managing Director, Digital Marketing Manager (continued) Table 1. Profile of the participant retailers Luxury fashion retailers’ postentry expansion 229

progressed to Stage Three only, and focused on national capital cities (tier-1) and regional and provincial capital cities (tier-2) for their physical stores. Two participant retailers adopted this strategy. Both entered China through high-cost organic growth. The second is regional expansion, through which the retailers have developed retail operations in a wider range of cities (including small number of tier-3 cities in affluent regions) stopping at Stage Four. Five participant retailers had adopted this strategy, and initially entered China through high cost organic growth. The third is countrywide expansion, through which the retailers are running Country of origin Retail format Ownership Year and city of China’s entry Entry mode E-commerce Number of interviews and interviewee position K UK Design house Private 1993 and Shanghai Organic growth Yes 2 and Retail Market Manager, Strategy Director L UK Design house Subsidiary 1997 and Beijing Licensing Yes* (through ecommercial platforms) 3 and Managing Director, Retail Director, ECommerce Manager M US Jewellery and watches Subsidiary 2001 and Shanghai Organic growth Yes 2 and Managing Director E-Commerce Manager N US Design house Private 1997 and Shanghai Franchising Yes 2 and Digital Market Manager, Table 1. Retail Director Initial entry market Post-entry expansion patterns A Beijing All tier-1 and tier-2 capitals, and small number of tier-3 cities in affluent regions B Beijing All tier-1 and tier-2 capitals, and small number of tier-3 cities in affluent regions C Beijing All tier-1 and tier-2 capitals, and large number of tier-3 cities in various regions D Shanghai All tier-1 and tier-2 capitals, and large number of tier-3 cities in various regions E Shanghai All tier-1 and tier-2 capitals, and small number of tier-3 cities in affluent regions F Beijing All tier-1 and tier-2 capitals, and large number of tier-3 cities in various regions G Beijing All tier-1 and tier-2 capitals, and large number of tier-3 cities in various regions H Shanghai All tier-1 and tier-2 capitals, and small number of tier-3 cities in affluent regions I Beijing All tier-1 and tier-2 capitals, and large number of tier-3 cities in various regions J Beijing All tier-1 national capitals and a few tier-2 regional capitals K Shanghai All tier-1 and tier-2 capitals, and small number of tier-3 cities in affluent regions L Beijing All tier-1 and tier-2 capitals, and large number of tier-3 cities in various regions M Shanghai All tier-1 national capitals and a few tier-2 regional capitals N Shanghai All tier-1 and tier-2 capitals, and large number of tier-3 cities in various regions Table 2. Post-entry expansion patterns of participant retailers IJRDM 49,2 230

Stage 1 Stage 2 Stage 3 Stage 4 Stage 5 Initial market entry through Beijing or Shanghai (standalone stores) All tier-1 national capital cities (standalone stores) Some tier-2 regional and provincial capital cities (standalone & flagship stores) More of tier -2 regional and provincial capital cities (standalone stores & flagship stores) & developing e- and m-commerce & optimising store profile and enhancing operations in national capital cities (standalone & flagship stores) Secondary cities (tier-3) in some regions (standalone stores & concessions) & strengthening e- and m-commerce Only applicable to some luxury fashion retailers Figure 2. Internationalising luxury fashion retailers’ post-entry expansion direction in China Luxury fashion retailers’ postentry expansion 231

retail stores in a large number of cities (more tier-3 cities in various regions). Half of the participant retailers had adopted this strategy. Five out of seven entered into China through medium cost local partnerships, including franchising and licensing. In Stage One, all participant retailers entered into China from national capital cities, either Beijing or Shanghai, through standalone stores in high-end hotels and luxury malls. Eight participant retailers who entered from Beijing developed their first retail stores in China at the Peninsula Hotel Beijing. Even today, luxury hotels remain important for luxury fashion retailers’ market entry in most of emerging and less developed countries, because they are capable of providing a safer investment environment and luxury shopping experience. Participant retailer B (a group owned French design house, entered China in 1992 through wholly owned standalone store in Beijing) justified such importance, and explained: Until today, high-end hotels are extremely important for any luxury brands in most emerging markets, because they are usually regarded as symbols of luxury, and they are really good at offering intangible value and luxury experiences for consumers there, as well as appropriate investment environment for us. During Stage Two, all of the participant retailers had developed their operations into the remaining tier-1 national capital cities, Guangzhou and Shenzhen, through standalone stores in luxury malls and hotels. The expansion in this stage was motivated by two main reasons. The first is that local knowledge about South China and experience obtained in established operations in Hong Kong prior to entering mainland China, proactively encouraged them to establish retail operations in both cities (Baiet al., 2018). The second is strongly related to their long-term growth strategy in China. In order to develop a strategic distribution network to cover most parts of the country, the retailers needed to establish operations in Guangzhou and Shenzhen (regional capital cities in South) after Beijing (North) and Shanghai (East). Participant retailerN(a private American design house, entered China from Shanghai in 1997 through a franchised standalone store) confirmed: Our business in China is a long-term strategy, we must establish a distribution network to cover all major regions, so we went to Guangzhou (South) right after Beijing and Shanghai. After establishing retail operations in all national capital cities, all participant retailers started to develop their business further into tier-2 capital cities in some provinces and autonomous regions in East, South and North in the Stage Three. Standalone stores remain popular and important, but flagship stores started to be adopted in some regional capital cities, such as Hangzhou (East). The two participant retailers, adopted cautious expansion strategies, halted further expansion before moving into the final two stages. The reasons include limited financial capacity because of expensive wholly owned expansion, and the desire to avoid diluting a highly exclusive brand image because of over distribution. For instance, participant retailer J (a private Swiss jewellery and watch specialist who entered China in 1999 through a wholly owned standalone store in Beijing) explained: As a top exclusive brand, difficult accessibility is vital for us. We do not want to dilute such brand image by over exposure... Directly owned operations are very expensive, we cannot afford to operate in many cities. During the Stage Four, 12 participant retailers further extended their operations into more tier-2 regional and provincial capital cities through standalone stores and a few flagship stores. Five participant retailers, who had adopted regional expansion strategies, have paused their offline expansion here. The fast-growing West China has also become increasingly important, for example Chongqing and Chengdu in Southwest and Xi’an in Northwest. Indeed, participant retailer A (a private French design house, entered China in 1996 through a wholly owned standalone store in Beijing) was preparing to open their first IJRDM 49,2 232

store in Northwest at Xi’an during this research. Meanwhile, the national capital cities became increasingly competitive, some participant retailers have started to strengthen their operations by running more standalone and flagship stores in Beijing and Shanghai. Nevertheless, four participant retailers indicated that overall post-entry expansion of physical stores had slowed down during this stage because of their desire to develop e-commerce and an increasingly saturated physical stores portfolio. Thus, they had started to develop their online operations through multiple channel strategies, and optimised their store profile through withdrawal from some less strategically important tier-2 provincial capital cities. For instance, participant retailer K (a private British design house, entered China in 1993 through a wholly owned standalone store in Shanghai) explained: We had established e-commerce through official online store (embedded in official website), and eflagship store at Tmall.com (a subsidiary of Alibaba Group). Such development helps to optimise store portfolio via opening more stores in strategically important cities and shutting stores in nonstrategically important cities. Only seven participant retailers who adopted countrywide expansion strategies had progressed to the final stage, through which they penetrated their offline operations further in to secondary cities (tier-3) in North, East and South. Indeed, due to rapid economic growth and fast urbanisation during the past decade, various regions, especially Pearl River Delta (South) and Yangtze River Delta (East), have become lucrative target markets. However, because the market infrastructure of these secondary cities is not as developed as capital cities, for example a lack of luxury malls, the retailers were compromised by local conditions and started to adopt concessions and standalone stores. Furthermore, not all of seven participant retailers followed a stepwise approach to reach the final stage. Three of them had achieved countrywide operations directly from their initial market entry. All of the three were private design houses, who entered China through local partnerships. Finally, as e-commerce has become increasingly important, participant retailers have start to strengthen their online performance through either multi-channel or omni-channel strategies (Bai, 2018). 4.2 The criteria for assessing local markets Alexander and Doherty (2009) identified a series of determinants which retailers adopt to assess foreign markets, including geographic and cultural proximity, similar economic development and social conditions, and similar or different market infrastructure. However, as suggested by this study, a different set of criteria should be considered for post-entry expansion, especially in China where considerable regional differences exist. Compared with supermarket chains, luxury fashion retailers have employed more complicated and interrelated criteria to assess local markets for post-entry expansion of retail operations. Firstly, location, strategic importance and market coverage (number of non-capital cities could be commercially covered) of a city are deemed top priority. This can be evidenced from indications that regional capital cities, such as Shenyang in Northeast and Zhengzhou in Central, all hosted expansion by large numbers of luxury fashion retailers. The benefits offered by these regional capital cities include solid economic growth, high levels of urbanisation, popular tourism spots, regional culture and extensive market coverage. Indeed, all participant retailers in this study, especially those who adopted cautious and regional expansion patterns, have extended their retail operations to these regional capital cities directly after their initial market entry. For instance, participant retailer H (a Spanish group owned design house, entered China in 2001 through a standalone store in Shanghai) highlighted such importance, and proposed: We must maintain our highly exclusive brand image through difficult accessibility in not only expensive price, but also limited distribution... We believe a city’s location and strategic importance, Luxury fashion retailers’ postentry expansion 233

such as market coverage, is one of the most factors used to assess potential market for physical stores, that is why we are running only three stores in West (Chengdu and Chongqing for Southwest, and Xi’an for Northwest). Secondly, similar to other types of retailers, luxury fashion retailers regard economic factors as important criteria to assess the potential of a target local market in China. The major indicators include current and potential economic growth and residents’ disposable income, and their willingness to pay. Indeed, in terms of the differences in economic development, over half of participant retailers prioritised tier-3 secondary cities in some wealthy regions to provincial capital cities in other less wealthy regions. Indeed, participant retailer G (a group owned Italian design house, entered China in 1993 through a franchised standalone store in Beijing) justified the importance of economic factors: Economy is paramount when we choose a local market in China, in Japan, everywhere... Besides the current economy and customers’ disposable income right now, we do value more potential growth and consumers’ willingness to buy in the future. Participant retailer A (a private French design house, entered China from Beijing in 1996 through a wholly owned standalone store) heavily relies on its customer relationship management system to measure consumer willingness to pay. The factors applied to assess willingness to pay include actual needs of local consumers, regional economic development, fashion trends, perceived product quality and value, as well as highly exclusive brand image. The Retail Operations Director stated that: We have developed a very good customer relationship management system, which helps us to obtain depth understanding of their real needs... Other factors considered (for willingness to pay) include brand image, perceived product quality and value, economic development, and fashion trend etc. Thirdly, due to the desire to strengthen brand awareness and raise brand loyalty through instore experiences, all of the participant retailers believed that their focus on post-entry expansion in China was still mainly about expanding retail stores during the past two decades. Therefore, from the perspective of retail store operation, appropriate store location and good quality staff are also important considerations. Additionally, the competition between luxury fashion retailers in order to obtain a good store location has led to increased costs, which is especially challenging for small-scaled private design house and/or new comers. This was confirmed by five out of seven participant retailers who adopted countrywide expansion strategies. For instance, participant retailer F (a private Italian design house, entered China in 1995 through a franchised standalone store in Beijing) explained: It is undoubtedly the case that national and regional capital cities (in China) have become mature markets. However, there is totally different story in some tier-2 and the majority of tier-3 cities, because of under developed market infrastructure and the challenge of recruiting good quality staff.... Gaining a good store location has become increasingly difficult in recent years. You know, customers will not regard your brand as a luxury brand if you do not have a store in ground floor of a prestigious mall. Fourthly, this study recognised that over half of participant retailers prioritised secondary cities (tier-3) to capital cities (tier-2) in some provinces due to a higher degree of urbanisation and the popularity of tourism. This is different from multinational supermarket chains who adopt the approach from provincial capital cities to the secondary cities (Wang and Li, 2012). For instance, participant retailer A explained: Qingdao is more suitable than Ji’nan because of its higher degree of urbanisation, more Western lifestyle, and greater popularity in terms of tourism, which can help us to attract new customers and even enhance the connection between us and loyal customers, so that we can provide better services in their own cities. IJRDM 49,2 234

Fifthly, although policies are usually regarded as one of the most important factors to assess a foreign market, they are not as significant for luxury fashion retailers’ post-entry expansion in China. This study identified two voices regarding the beneficial policies and the favourable offers provided by local government and landlords. These favourable offers, such as reduced taxation and rents, are very attractive for small-scale private design houses, especially those who had entered and expanded through local partnerships. Six participant retailers thought that this may help them to obtain a good store location in some cities, where competition is not yet high. On the other hand, other participant retailers, especially group owned subsidiaries, did not believe such favourable policies and offers are sustainable enough for their long-term business development strategies due to the risks to dilute exclusive brand image through over distribution in non-key cities. Therefore, they did not want to expand into somewhere not considered as appropriate markets for their brands whatever policies and/or offers are available. For instance, three participant retailers had withdrawn from a few tier-2 provincial capital cities, such as coal mining cities in North, because they had realised their brand image could be diluted through over expansion, and repetition of market coverage in regional capital cities and even through e-commerce. Finally, the rapid development of e-commerce and interactive direct marketing during the last decade have made a great impact on participant retailers’ operation markets and brick-and-mortar store portfolios. Therefore, e-commerce has also become an important factor for assessing a target local market in China. This effect will be discussed in depth in the following section. 4.3 The impact of e-commerce on post-entry expansion direction Bai (2018) argued that fast growing e- and m-commerce have made a great impact on luxury fashion retailers’ internationalisation and post-entry expansion strategies. Indeed, thirteen out of the fourteen participant retailers have developed e- and m-commerce in China in recent years, through various of online channels. Moreover, this research identified two areas of impact of e- and m-commerce on luxury fashion retailers’ post-entry expansion direction in terms of physical store profile and retail store operation cities. On the one hand, four participant retailers adopted multi-channel distribution strategies, those of e- and m-commerce as additional channels to distribute. Indeed, they had adopted as many online channels as possible, including wholly owned official online stores, social media and e-commerce platforms, as well as online luxury retailers. All of the four participants, adopted a countrywide expansion direction pattern, have penetrated further into a wider range of local markets as well as e- and m-commerce. For instance, besides rapidly expanding e-commerce through official online store, e-flagship stores at JD.com and T-mall, and m-commerce through Chinese social media, participant retailer N had increased the numbers of brick-and-mortar stores in over 50 cities, and said: As a lifestyle luxury fashion brand, we aim to reach as many consumers as we can. I believe that the rapid growing e-commerce helped us to identify increasing numbers of potential cities for retail stores. On the other hand, nine participant retailers who had adopted omni-channel distribution strategies, regarded e- and m-commerce as a part of their holistic distribution systems. They therefore aim to achieve the most efficient and effective distribution systems through integrating both e- and m-commerce and brick-and-mortar stores. Similar to the four participants who adopted multi-channel countrywide expansion direction strategies, the participant retailers who employed omni-channel countrywide and/or (for some) regional expansion patterns were also keen to develop physical stores in more local markets, especially where e-and m-commerce are relatively developed. For instance, participant retailer C (a group owned German accessories and leather goods specialist, entered China in Luxury fashion retailers’ postentry expansion 235

1996 through a wholly owned standalone store in Beijing) explained that they have penetrated their retail stores further into more tier-3 cities in some affluent regions because of the strong performance of e- and m-commerce in those regions, particularly East and South. However, other participant retailers, who adopted omni-channel (for some) regional and/or cautious expansion patterns, are more likely to reduce the numbers of physical stores. One reason is the desire to protect their highly exclusive brand image from over distribution. They therefore employed e
and m-commerce as a way to strengthen commitment to consumers through more integrated services and brand experiences. They thus had increased investment in logistics and information systems and had started to offer “Click and Collect” services. The other reason was the desire to optimise their physical store profile. Then they are able to allocate resources to more efficient and effective markets. For example, in parallel to developing the official online store within an omni-channel strategy, participant retailer B had optimised distribution channels by shutting stores in some less strategically important cities but opening new stores in key cities, and explained: China has become a competitive market, it is necessary for us to offer a unique and seamless brand experience via an omni-channel strategy... Such a strategy has helped us to allocate resources more efficiently. For instance, we have closed a store in Taiyuan (a tier-2 provincial capital in North), and opened more stores in Beijing. In summary, based upon the findings in this section, it is known that all national capital cities, most of regional capital cities, and increasing numbers of secondary cities in the East, North and South, especially where strong performing e- and m-commerce exist, are increasingly important markets for luxury fashion retailers’ post-entry expansion of their brick-andmortar stores within the digital era. All of the three sections of research findings presented in this chapter will be discussed further in the following chapter. 5. Discussion Here are several important implications for the initial entry into the Chinese market. Firstly, because of the considerable perceived psychic distance and the experience gained from Hong Kong, fashion retailers are more likely to adopt the entry modes which require high and medium costs and provide high and medium levels of control, such as organic growth, franchising and licensing to seek local partnerships. This is different from the entry modes adopted to enter into traditional luxury fashion markets, where expensive flagship stores (Moore et al., 2010) and low cost export, wholesale and concessions (Moore et al., 2000) are popular. Therefore, in the context of emerging market, high cost entry modes are not necessarily associated with high risk, because these expensive strategies offer the retailers higher degree of decision-making and control over their brands and business activities abroad (Bai et al., 2017). Secondly, due to significant differences in terms of market infrastructure in China such as lack of high-end department stores and premium shopping districts, luxury fashion retailers are keen to adopt standalone stores in luxury malls and hotels, rather than concessions. Even in the contemporary digital era, e-commerce is still not a popular method used by luxury fashion retailers for their initial retail operations in China because in-store services, such as interpersonal communication, is a critical component of the intangible value of any luxury fashion brand (Kapferer, 2015). Thirdly, there is only one choice for initial entry into some developed markets, such as London for the UK However, due to considerable intrinsic regional differences and even geographic location, Beijing and Shanghai and even Hong Kong are widely regarded as the entry city of (Greater) China (Bai et al., 2018). This can be evidenced by the fact that the retailers extended their retail operations in Guangzhou and Shenzhen (regional capital cities in South). IJRDM 49,2 236

The important implications for luxury fashion retailers’ post-entry expansion direction in China include the following. Firstly, according to the scope of operation markets and physical store profiles, three kinds of post-entry expansion direction patterns have been identified: namely cautious expansion, regional expansion and countrywide expansion. This is different from developed but small geographic markets, because the retailers have stopped further penetration beyond regional capital cities (Moore et al., 2000). Therefore, luxury fashion retailers can tailor their post-entry expansion in terms of number, format and location of retail stores as well as e- and m-commerce, according to their strategies regarding to branding and business development. Secondly, many luxury fashion retailers still adopt stepwise strategies (from tier-1 national capital cities to tier-2 regional and provincial capital cities and tier-3 secondary cities in any provinces) to expand further within China, which is similar as generic retailers and supermarket chains (Wang and Li, 2012). However, the tier system of China’s cities has been more or less weakened, because some luxury fashion retailers have started to prioritise a few tier-3 cities in affluent provinces and regions over tier-2 capital cities in less affluent provinces and regions. Thirdly, the overall direction of luxury fashion retailers’ post-entry expansion in China is from Central and South to North, and from East to West. Moreover, expansion of intra- and inter-regions can be simultaneous. Fourthly, luxury fashion retailers further expanded their retail stores in a wide range of local markets in China through their owned or local partners’ resources rather than through the networks of distribution retailers, such as department stores. This is different from in developed markets, such as the UK wherein luxury fashion retailers can extend from London to Edinburgh through Harvey Nichols (Chevalier and Gutsatz, 2020). Fifthly, various store formats have been adopted for post-entry expansion in China. Particularly flagship stores that have become increasingly important for post-entry expansion because they help the retailers to strengthen operations in national capital cities and other key markets as a business development strategy, rather than initial market entry in emerging markets (Bai et al., 2017). Finally, in order to secure sustainable success, luxury fashion retailers assess and select local markets in China through various interrelated criteria, including location and strategic importance (geographic market coverage) of a city, economic development, availability of store locations and staff, a high degree of urbanisation and tourism, debatable favourable policies and offers (offered by local government and landlords), as well as the popularity of e-and m-commerce. The impact of e- and m-commerce on luxury fashion retailers’ brick-and-mortar store profiles and local markets are different, and depend on the retailers’ branding strategies, distribution strategies and sustainable business development outlook (Beck and Rygl, 2015). Indeed, some lifestyle luxury fashion retailers are keen to adopt multi-channel distribution strategies due to their desire to reach as many consumers as possible through countrywide expansion patterns and by expanding their retail facilitates in increasing numbers of local markets. On the other hand, the retailers who possess a highly exclusive brand image and who have expanded through a cautious expansion pattern, are more likely to adopt omni-channel distribution strategies. The motives for this include the ambition to offer constant and seamless brand experiences (Fisher et al., 2019), and the desire to gain the most efficient and effective performance by optimising typically reducing numbers of retail stores and less strategically important cities (Ailawadi and Farris, 2017). Furthermore, retailers who expanded further through regional expansion patterns behaved differently because they could increase or reduce the number of retail stores and operation markets through either multi- or omni-channel distribution strategies. Therefore, there is a positive relationship between the popularity of e- and m-commerce in a city and the potential of that city to be targeted as an operational market for luxury fashion retailers’ brick-and-mortar stores. Luxury fashion retailers’ postentry expansion 237

6. Conclusion From the perspective of international retailing, this study examines the direction of retailers’ post-entry expansion after initial entry of a host market within the context of luxury fashion market in China. Since this topic has not been previously considered, this paper offers depth insights for the current literature in the field of international retailing. It has not only examined luxury fashion retailers’ post-entry expansion direction patterns through operational details model, but has also identified the criteria used to assess a local market, and the impacts of e- and m-commerce towards retail store profile and operation markets. Moreover, the present study highlights the differences in relation to retail market infrastructures between developed and emerging markets, and stresses the importance of luxury malls and high-end hotels for luxury fashion retailers’ brickand-mortar stores. The results of this paper provide practitioners understanding to succeed in the international market through selecting appropriate entry mode, entry cities and local market assessment criteria for post-entry expansions. These insights are valuable for young and/or small-scale private luxury fashion retailers who are keen to internationalise into emerging markets. The validity and reliability of this study have been strengthened through using triangulation of qualitative data. This has been achieved through multiple interviews from different senior management within the same participant retailers and marketing reports. Nevertheless, the limitations of this study are considered in two aspects. The first is relation to the marketing context. The research findings are generated from China. They are thus probably limited in value to explain luxury fashion retailers’ post-entry expansion direction patterns elsewhere, where there are large differences in terms of policies, economic development, social and cultural conditions, and even geographic size exist. The second is involved to the relatively small sample size. However, 30 interviews from fourteen respondent retailers, across a wide range of retailing formats, ownership structures, country of origin and expansion strategies, are perceived to be strong enough to represent actuality in the market. Future studies are suggested to extend this study’s findings through examining other luxury fashion retailers who entered a host market and extended their operations further through e-commerce only. The researchers also may test the updated stepwise model developed in this study through quantitative studies by large numbers of samples to examine the similarities and differences across different market environment. References Ailawadi, K. and Farris, P. (2017), “Managing multi- and Omni-Channel distribution: metrics and research directions”, Journal of Retailing, Vol. 93 No. 1, pp. 120-135. Alexander, N. and Doherty, A. (2009), International Retailing, OUP, Oxford. Arora, A., McIntyre, J., Wu, J. and Arora, A. (2015), “Consumer response to diffusion brands and luxury brands: the role of country of origin and country of manufacture”, Journal of International Consumer Marketing, Vol. 27 No. 1, pp. 3-26. Bai, H., McColl, J. and Moore, C. (2017), “Luxury retailers’ entry and expansion strategies in China”, International Journal of Retail and Distribution Management, Vol. 45 No. 11, pp. 1181-1199. Bai, H., McColl, J. and Moore, C. (2018), “Hong Kong, a Gateway for Mainland China? An examination of the impact of luxury fashion retailers’ ownership structures on expansion strategies”, International Journal of Retail and Distribution Management, Vol. 46 No. 9, pp. 850-869. Bai, H. (2018), “Luxury fashion retailers’ Omni-Channel distribution and communication strategies in mainland China”, in Teller, C., Brusset, X. and Kotzab, H. (Eds), CERR 2018: Colloquium on European Research in Retailing, University of Surrey, pp. 120-124. IJRDM 49,2 238

Bain & Co (2017), “2017 China luxury market study – China setting luxury trends”, available at: http:// www.bain.com.cn/pdfs/201801180441238002.pdf. Bain & Co (2019), “Eight themes that are rewriting the future of luxury goods”, available at: https:// www.bain.com/globalassets/noindex/2020/bain_digest_eight_themes_that_are_rewriting_the_ future_of_luxury-goods.pdf. Beck, N. and Rygl, D. (2015), “Categorization of multiple channel retailing in Multi-, Cross-, and OmniChannel Retailing for retailers and retailing”, Journal of Retailing and Consumer Services, Vol. 27, pp. 170-178. Boisvert, J. and Ashill, N. (2018), “The impact of branding strategies on horizontal and downward line extension of luxury brands: a cross-national study”, International Marketing Review, Vol. 35 No. 6, pp. 1033-1052. Chen, J., Teng, L., Liu, S. and Zhu, H. (2015), “Anticipating regret and consumers’ preferences for counterfeit luxury products”, Journal of Business Research, Vol. 68 No. 3, pp. 507-515. Chevalier, M. and Gutsatz, M. (2020), Luxury Retail and Digital Management: Developing Customer Experience in a Digital World, 2nd ed., John Wiley & Sons, Singapore. Chevalier, M. and Lu, X. (2009), Luxury China: Market Opportunities and Potential, John Wiley & Sons, Singapore. Christmann, H., Alexander, A. and Wood, S. (2016), “Exploring brand identity and entrepreneurship as drivers of small specialist retailer internationalisation: a German case study”, International Review of Retail, Distribution and Consumer Research, Vol. 26 No. 2, pp. 137-153. Creswell, J. and Creswell, J. (2018), Research Design: Qualitative, Quantitative and Mixed Methods Approaches, 5th ed., SAGE, Los Angeles. Deloitte (2017), “Global powers of luxury goods 2017 – the new luxury consumer”, available at: https:// www2.deloitte.com/content/dam/Deloitte/global/Documents/consumer-industrial-products/gxcip-global-powers-luxury-2017.pdf. Evans, J., Mavondo, F. and Bridson, K. (2008), “Psychic distance: antecedents, retail strategy implications, and performance outcomes”, International Marketing Review, Vol. 16 No. 2, pp. 32-63. Fisher, M., Gallino, S. and Xu, J. (2019), “The value of rapid delivery in omnichannel retailing”, Journal of Marketing Research, Vol. 56 No. 5, pp. 732-748. Frasquet, M., Dawson, J. and Alejandro, M. (2013), “Post-entry internationalisation activity of retailers: an assessment of dynamic capabilities”, Management Decision, Vol. 51 No. 7, pp. 1510-1527. Frasquet, M., Dawson, J., Calderon, H. and Fayos, T. (2018), “Integrating embeddedness with dynamic capabilities in the internationalisation of fashion retailers”, International Business Review, Vol. 27 No. 4, pp. 904-914. Gomes, R., Carneiro, J. and Dib, L. (2018), “Branded retailer expansion on a continent-sized emerging market”, International Journal of Retail and Distribution Management, Vol. 46 No. 9, pp. 820-834. Guercini, S. and Milanesi, M. (2017), “Extreme luxury fashion: business model and internationalization process”, International Marketing Review, Vol. 34 No. 3, pp. 403-424. Hennigs, N., Wiedmann, K., Klarmann, C. and Behrens, S. (2015), “The complexity of value in the luxury industry: from consumers’ individual value perception to luxury consumption”, International Journal of Retail and Distribution Management, Vol. 43 Nos 10/11, pp. 922-939. Hutchinson, K. and Quinn, B. (2011), “Identifying the characteristics of small specialist international retailers”, European Business Review, Vol. 23 No. 3, pp. 825-835. Kapferer, J. and Valette-Florence, P. (2016), “Beyond rarity: the paths of luxury desire. How luxury brands grow yet remain desirable”, Journal of Product and Brand Management, Vol. 25 No. 2, pp. 120-133. Kapferer, J. (2015), “The future of luxury: challenges and opportunities”, Journal of Brand Management, Vol. 21, pp. 716-726. Luxury fashion retailers’ postentry expansion 239

Ko, E., Costello, J. and Taylor, C. (2019), “What is a luxury brand? Anew definition and review of the literature”, Journal of Business Research, Vol. 99, pp. 405-413. Laforet, S. and Chen, J. (2012), “Chinese and British consumers’ evaluation of Chinese and international brands and factors affecting their choice”, Journal of World Business, Vol. 47 No. 1, pp. 54-63. Liu, X., Burns, A. and Hou, Y. (2013), “Comparing online and in-store shopping behaviour towards luxury goods”, International Journal of Retail and Distribution Management, Vol. 41 Nos 11/12, pp. 885-900. Liu, S., Perry, P., Moore, C. and Warnaby, G. (2016), “The standardisation- localisation dilemma of brand communications for luxury fashion retailers’ internationalisation into China”, Journal of Business Research, Vol. 69 No. 1, pp. 357-364. Lopez, C. and Fan, Y. (2009), “Internationalisation of Spanish fashion brand Zara”, Journal of Fashion Marketing and Management, Vol. 13 No. 2, pp. 279-296. Lu, X. (2012), “Market entry strategies of international luxury brands in China”, Finance and Trade Economics, Vol. 33 No. 8, pp. 105-112. Makkar, M. and Yap, S. (2018), “Emotional experiences behind the pursuit of inconspicuous luxury”, Journal of Retailing and Consumer Services, Vol. 44, pp. 222-234. McKinsey & Co (2017), “2017 China Luxury Report – Chinese luxury consumers: the 1 trillion renminbi opportunity”, available at: https://www.mckinsey.com/∼/media/McKinsey/BusinessFunctions/ MarketingandSales/OurInsights/ChineseluxuryconsumersMoreglobalmoredemandingstillspending/ Chinese-luxury-consumers-the-1-trillion-renminbi-opportunity.ashx. Moore, C., Fernie, J. and Burt, S. (2000), “Brands without boundaries – the internationalisation of the designer retailer’s brand”, European Journal of Marketing, Vol. 34 No. 8, pp. 919-937. Moore, C., Doherty, A. and Doyle, S. (2010), “Flagship stores as a market entry method: the perspective of luxury fashion retailing”, European Journal of Marketing, Vol. 44 Nos 1/2, pp. 139-161. Ojala, A. (2015), “Geographic, cultural and psychic distance to foreign markets in the context of small and new ventures”, International Business Review, Vol. 24 No. 5, pp. 904-914. Schu, M. and Morschett, D. (2017), “Foreign market selection of online retailers – a path-dependent perspective on influence factors”, International Business Review, Vol. 26 No. 4, pp. 710-723. Shukla, P., Banerjee, M. and Singh, J. (2016), “Customer commitment to luxury brands: antecedents and consequences”, Journal of Business Research, Vol. 69 No. 1, pp. 323-331. Siebers, L. (2017), “Hybridization practices as organizational response to international demands: the development of Western retail TNCs in China”, Journal of Economic Geography, Vol. 17 No. 1, pp. 1-29. Spiggle, S., Nguyen, H. and Caravella, M. (2012), “More than fit: brand extension authenticity”, Journal of Market Research, Vol. 49 No. 6, pp. 967-983. Swoboda, B., Zentes, J. and Elsner, S. (2009), “Internationalisation of retail firms: state of the art after 20 years of research”, Marketing: Journal of Research and Management, Vol. 5 No. 2, pp. 105-126. Vahlne, J. and Johanson, J. (2017), “From internationalization to evolution: the Uppsala model at 40 years”, Journal of International Business Studies, Vol. 48 No. 9, pp. 1087-1102. Wang, X. and Li, Y. (2012), “Foreign retailers’ localization strategies in mainland China – the cases of Wal-Mart, Carrefour and Metro”, Journal of Beijing Technology and Business University: Social Science, Vol. 3, pp. 34-41. Wang, X. and Liu, Y. (2008), “An empirical study on the location selection of foreign direct investment in China’s retailing industry”, China Business and Market, Vol. 21 No. 6, pp. 57-60. Ye, Y., Lau, H. and Teo, L. (2018), “Drivers and barriers of omni-channel retailing in China: a case study of the fashion and apparel industry”, International Journal of Retail and Distribution Management, Vol. 46 No. 7, pp. 657-689. IJRDM 49,2 240

Yin, R. (2018), Case Study Research and Applications: Design and Methods, 6th ed., SAGE, Los Angeles. Yu, W. and Ramanathan, R. (2012), “Effect of business environment on international retail operations: case study evidence from China”, International Journal of Retail and Distribution Management, Vol. 40 No. 3, pp. 218-234. Corresponding author Huifeng Bai can be contacted at: [email protected] For instructions on how to order reprints of this article, please visit our website: www.emeraldgrouppublishing.com/licensing/reprints.htm Or contact us for further details: [email protected] Luxury fashion retailers’ postentry expansion 241

Lean thinking in the luxury-fashion market Evidences from an extensive industrial project Gionata Carmignani Department of Energy, Systems, Territory and Constructions, University of Pisa, Pisa, Italy, and Francesco Zammori Department of Industrial Engineering, University of Parma, Parma, Italy Abstract Purpose – The capability to overcome tradeoffs among costs, quality and time has become a must in high-margin businesses too. Lean thinking may be a solution, but applications in the luxury-fashion market are still rare. In order to shed light on this apparent contradiction, the purpose of this paper is to identify the key features of the luxury-fashion market that may act as barriers for the adoption of lean principles. Next, based on the results of this preliminary analysis, the paper tries to verify, if and how, lean principles can be properly reinterpreted, so as to properly fit the requirements of this market. Design/methodology/approach – Due to the operating nature of lean, an empiric approach was followed. From the evidences gathered during a lean project of a world-wide company, critical elements of the luxury-fashion market were identified and used as criteria to select, among lean tools, the most appropriate ones. Lastly, selected tools were integrated in a structured framework (for lean implementation) that was used to analyze and to improve many logistics and manufacturing processes. Findings – Developed solutions were implemented as pilot projects, with outstanding preliminary result. Results are case specific and trying to infer general considerations may be hazardous. Nonetheless, due to the relevant dimension of the project, they can be considered more than a clue concerning the robustness of the framework and, most of all, concerning the real potentialities of lean in the luxury-fashion market. Practical implications – The framework is extremely operational and, together with the proposed industrial cases, can be used as a guideline to support practitioners during the implementation of similar projects. Originality/value – Lean thinking is relatively new in the luxury-fashion market, where the focus on operational costs has been traditionally considered as a marginal issue. Thus, the application of lean principles in this market is the innovative element of the paper. Keywords Lean thinking, Value stream mapping, Lean toolkit, Luxury-fashion market, Visual management Paper type Case study 1. Introduction Due to new technologies and globalization, satisfying customers is more and more challenging and, in order to be competitive in an ever changing marketplace, enterprises are urged to rethink their business models and to decline it from the strategic to the operational level (Voss, 1995; Li and Tan, 2004; Raymond and Croteau, 2006). In a word, the capability to overcome the traditional tradeoffs among costs, quality and time (i.e. flexibility) has become a must (Bertolini et al., 2013). In this respect, new managerial paradigms, such as world class manufacturing (WCM), have emerged as effective competitive weapons for several businesses, in that they represent a synergic combination of lean practices acting as the pillars of an integrated International Journal of Retail & Distribution Management Vol. 43 No. 10/11, 2015 pp. 988-1012 © Emerald Group Publishing Limited 0959-0552 DOI 10.1108/IJRDM-07-2014-0093 Received 10 July 2014 Revised 4 December 2014 27 April 2015 Accepted 6 May 2015 The current issue and full text archive of this journal is available on Emerald Insight at: www.emeraldinsight.com/0959-0552.htm 988 IJRDM 43,10/11

systems (Hines et al., 2004; Braglia et al., 2009) that encompasses all the areas of a business, from the strategic to the operating level (Schonberger, 1996; Yamashima, 2000). However, most of the times the core of WCM focusses on internal operations (i.e. logistics, production, maintenance and quality assurance) and, in this sense, it can be considered as a synonym of lean manufacturing. Specifically, from this perspective, the main goal is to match and even to outdo customers’ needs and expectations, through a flexible and resilient manufacturing system focussed on waste removal and value creation (Bonaccorsi et al., 2011). The main idea is to preserve value with less work, through the identification and progressive removal of all non-value-adding (NVA) activities, or muda, defined as “any use of resources that is not aimed at the generation of value” (Womack and Jones, 1996). Continuous flow is another key concept of lean, but it is also the hardest element to be understood and achieved: after muda have been eliminated, work should be able to flow, constantly and smoothly, from a value added (VA) activity to the next one, without obstacles and/or barriers. This state represents the ideal value stream and should be the target of any world class manufacturer. Benefits of lean are well known and have been demonstrated by many successful applications in several industrial fields (Kuldip and Sangwan, 2014), which had led to astonishing results in terms of lead time reduction and stabilization, process simplification, inventory reduction, quality improvement, people empowerment and satisfactions (Melton 2005; Demeter and Matyusz, 2011). Also, and perhaps more important, although lean originated for “high-volume low-variety” manufacturing firms, this managerial philosophy has been successfully applied even in services and process industries (Rahani and Al-Ashraf, 2012; Nugroho et al., 2012; Aziz and Hafez, 2013; Bhamu et al., 2013). Nonetheless, lean should not be considered as a straightforward solution for every problem; conversely, the route toward an ideal value stream is long and running into failures is quite easy (Alvarez et al., 2009). In this respect, the point is that what distinguish world class companies is not the use of spectacular concepts, but rather it is the unconventional (i.e. innovating) way in which known tools and methods are re-interpreted and applied. It is precisely the need to re-interpret and to decline lean principles into operations that makes the development process challenging and hard to be generalized (Bertolini et al., 2013). As a matter of fact a universal framework does not exists and the development process requires both wit and expertise, as testified by many works that tried to tailor lean to the requirements and to the specificities of different market segments (Muda and Hendry, 2003; Achanga et al., 2006; Alvarez et al., 2009). The present paper belongs to this stream of research and addresses lean in the luxury-fashion market. More specifically, the focus is on Italian luxury fashion goods that represent the cornerstone of “made in Italy” and that account for more than 24 percent of the overall luxury goods sold worldwide (D’arpizio, 2013). This market is constantly growing, between 4 and 6 percent a year, and has a high profitability with an average net profit margin of 8.5 percent and a return on asset of 7.4 percent (Arienti, 2013; D’Arpizio, 2014). Nonetheless, due to the recent economic crisis, the capability to overcome tradeoffs among costs, quality and time has become a must in this market too, since customers can be easily lost, unless costs are cut without affecting neither quality nor service level. Lean thinking may be a solution, but its implementation is challenging and applications in the luxury-fashion market are still very rare (Olivera et al., 2007). Concerning the last assertion one could argue that lean cannot be considered as a totally new concept for this market, since the first applications can be traced back to the late 1990s when the concept of agile and leanagile supply chain was first introduced (Christopher and Peck, 1997). However, most of 989 Lean thinking in the luxury-fashion market

the industrial applications were made at a strategic and tactical level rather than at the operating one; in other words lean has been extensively applied in terms of supply chain management (Christopeher et al., 2004; Masson et al., 2007), but applications concerning internal operations are still uncommon (Bruce et al., 2004; Qi et al., 2007; Bhardwaj and Fairhurst, 2010). There is no doubt that an excellent supply chain is a key element of success, in a market with rapid and unpredictable dynamics (Green et al., 2014) and, in this sense, lean-oriented concepts such as supply chain’s resilience, supply chain’s agility and total JIT (i.e. an integrated supply chain strategy incorporating JIT-purchasing, JIT-selling and JIT-information) are essential in the luxury fashion industry. However, our belief is that the application of lean at the strategic and tactical level is not enough and that lean should be exploited also at the operation level, both in logistics and manufacturing processes. This is because, as for all high margin-businesses, also in this sector the focus on operational costs has been traditionally considered as a marginal issue and so it is still common to find “old fashioned” manufacturing systems, whose inefficiencies have long been concealed by stocks and where there is ample space for improvement. Certainly, extending lean also to the operating processes may be challenging, because many specificities of the luxury-fashion industry (such as a high variability and a high level of product customization) could act as barriers for lean initiatives. Nonetheless, recent successful projects, in services industries and public services, suggest that these problems can be solved also in those sectors that had always been considered incompatible for lean (Arlbjørn et al., 2011). In virtue of the above mentioned issues, the objective of the paper is twofold: to understand which critical issues may limit the adoption of lean in the luxury-fashion industry; and to clarify if and how these barriers can be broken down, by properly reinterpreting lean principles in an innovative and/or alternative way. Due to the absence of adequate bibliographic sources and of significant industrial applications, neither a literature research nor a survey approach was feasible; thus, in order to give a preliminary answer to our research questions, an empirical approach was followed. Specifically, moving from the expertise acquired during an extensive lean project (brought about by a world-wide operating company), a framework for lean implementation was formulated, using as main pillars a sub-set of lean tools, which resulted more aligned with the specificities of the investigate market segment. The rest of the paper is organized as follows. Section 2 pinpoints the main elements of the proposed framework and shows how it is linked to the specificities of the luxury-fashion company for which it was conceived. Next, two relevant industrial cases, which detail how the framework was actually used to improve manufacturing and logistics processes, are described in Sections 3. Lastly, conclusions and future research directions are drawn in Section 4. 2. A framework for lean introduction in the luxury fashion industry As noted in the Introduction, a general framework for lean implementation does not exist. Although a comprehensive set of lean tools has been widely described in technical literature (Hines et al., 2004) the majority of them can be applied only in specific sectors or, at least, they need being reconfigured depending on the business and market conditions, product and company characteristics (Turner and Auer, 1996; Lewis, 2000). Also, and perhaps more important, the luxury-fashion industry is one of the market where lean could be more difficult to implement, due to a series of critical elements that will be detailed later on. For this reason, in order to properly 990 IJRDM 43,10/11

contextualize lean tools to the specific requirements of the investigated market segment, we decided to follow an empirical approach. Specifically, the framework for lean implementation in the luxury-fashion industry was developed on the field, during a lean project that involved several facilities owned by a famous Italian fashion brand, whose name cannot be revealed for reason of confidentiality and that, hereafter, will be referred as Italian Fashion Company (IFC). 2.1 IFC’s main figures and objectives of the lean project Born as a fur coats workshop in the early 1950s, IFC has rapidly grown, moving from a quasi-artisan family business to a multinational company rated as one of the top ten luxury fashion brands in the world. While still maintaining an organizational structure that reflects a centralized hierarchical-functional frame, nowadays IFC operates worldwide with an average yearly revenue of €800 million. Europe is still the main market, but exports to East and Far East countries are constantly increasing, with a share of the total sales higher than 35 percent. IFC sells also in North America and Mexico, but in these countries the sales network is less pervasive and this market roughly corresponds to 15 percent of the total sales. The product range is wide and the main products are clothing, furs, bags, accessories, perfumes and shoes. The market idea pursuit by the brand is that “to sell an emotion” not only a cloth or an accessory. Thus, all products, while not unique, are produced in limited quantities with a “tailor made feeling,” so as to transmit an idea of elegance, culture of design, Italian tradition, heritage and style. High-quality standards, a continuous innovation of both style and materials and the “Made in Italy Identity” are also fundamental to transfer this market idea. This is the reason why both production and design are performed in Italy: new collections of ready to wear and off-the-peg items (i.e. non-clothing such as handbags) are continuously created in the headquarter located in Rome, and finished goods are assembled and/or tailored in 12 manufacturing facilities, owned by IFC, and positioned in the proximity of the headquarter. However, moving upstream the supply chain, the structure of the in-bund network branches and greatly expands; indeed, for fabrics, leathers and precut components multi-sourcing and/or subcontracting are frequently used. The sense of uniqueness and the excellence of the brand are reinforced by a selective distributions network, based on more than 140 mono-brand stores located in exclusive and trendy neighborhoods of the major cities in and outside Europe. Italian stores are directly replenished by a central warehouse located in Rome, whereas the other ones are replenished by 15 local warehouses positioned all over the world (ten in Asia, three in Europe and two in the USA). IFC’s sales volumes are still growing (between 2 and 3 percent a year), but this is mainly due to exports to the East and Far East markets, especially Russia and China. The situation is different in EU countries where the market is shrinking. Specifically, due to margins contraction and high logistics costs, inventory reduction and rapidity are emerging as new vital objectives; a situation that prompted IFC to rethink its business processes. Due to the rapidity of the fashion cycle, the main problems concern obsolete materials and/or unsold products, two issues that have a high negative incidence on profitability. Thus, the operating goal is to reduce lots size, cut lead time and increase flexibility so as to postpone the launch of new collections as late as possible, avoiding “blind orders” of fabrics, leathers and other valuable materials. To this aim, at first, IFC tackled the problem focussing mainly on the outbound supply chain; in this respect several interventions have been made, the main ones being the 991 Lean thinking in the luxury-fashion market

adoption of Virtual Pooled Inventory and Distribution Requirement Planning. Nonetheless, to boost results, IFC has recently realized the need to move to the operating level, so as to improve performance of internal operations, too. This led to the development of a lean project that, under the supervision of a team of academic experts and top managers, has tackled the criticalities of several IFC’s facilities and logistic hubs, with the objective to cut operating costs and inventory levels. 2.2 Development of a conceptual framework: the starting point of the lean project Before moving to the operational level, in agreement with IFC’s top management, we decided to define a step by step procedure (i.e. a framework) for waste analysis and lean implementation, to be used in all subsequent tasks of the lean project. It is worth noting that, as a secondary objective we also tried to envisage the lean tools more likely to be used, in order to train (on these specific tools) the staff involved in the project. Thus, the first step of the analysis was to identify and to match the IFC’s business specificities with the lean tools described in technical literature (Osada, 1991; Hirano, 1995; Womack and Jones, 1996; Chen and Meng, 2010). As known, although the concept of luxury is characterized by a multidimensional structure of meanings particularly difficult to define (Wiedmann and Hennigs, 2013), there are a number of key elements to maintain a luxury fashion brand: clear brand identity, marketing communications, product integrity, design signature, high price, exclusivity, heritage, service and culture (Brun et al., 2008; Kapferer and Bastien, 2009). From these key elements descend a series of recurring business’ specificities (Caniato et al., 2009, 2011, Wiedmann and Hennigs, 2013), which, in the case of IFC are the following ones: (1) long and wide inbound supply chain (L_SC); (2) frequent use of subcontracting (Sb_C); (3) high demand variability (H_Dv); (4) high product customization and even product’s uniqueness (H_Pc); (5) short products’ life cycle (S_Pl); (6) high quality rate (H_Qr); and (7) products that are often the results of production processes that require artistic-craft (A_Cr). How can be easily understood, all these elements increase the complexity of the reference market and, above all, demand seasonality and the extensive number of product variants may hinder the adoption of many operational lean techniques which, as known, originated for low-variability high-volume productions with stable demand (Bruce et al., 2004; Holweg, 2007). Obviously, business’ specificities do not pose serious constraints to the introduction of “high level” managerial tools related to the corporate culture (leadership, training, communication, etc.). Conversely, the situation becomes much more complicated for low-level techniques operating at the shop floor level. In this case, only “visual-oriented tools” can be used straightforwardly, because these tools, although operating, make leverage on general concepts such as order, tidiness and adoption of clear and driving key performance indicators. Vice versa, for the other operating tools a definite answer cannot be given (Boyle and Scherrer-Rathje, 2009): we are not saying that these tools cannot be 992 IJRDM 43,10/11

used, but certainly their adoption require a more careful evaluation and, as we will show in Section 3, these tools should be rethought and properly redesigned (Hodge et al., 2011). More specifically, JIT-oriented tools are certainly the most critical ones because, due to high customization and demand variability a job shop with ever changing activities and unpredictable routings is the typical manufacturing environment. Thus, achieving a full process standardization and a leveled and paced manufacturing flow may be difficult, as production sequence is never stable and levelled sequences cannot be used (Hüttmeir et al., 2009). Similarly, an extremely short products life cycle implies unstable production mixes and the need for frequent layout’s readjustments, elements that are barriers for mixed-model production and cellular manufacturing. In this respect, collections’ seasonality is another huge problem, as it almost prevents the possibility to take advantage of processes commonalities. Indeed, each year at least two collections (for the winter and summer season) must be proposed, but, due to the very long lead time of the production/supply chain, all the activities of the manufacturing process (from design to delivering) must be strongly anticipated (even more than a year), with respect to the market launch. In addition to this, also workforce’s craftsmanship increases the complexity of the luxury-fashion market. On the one side this is the key element to assure products appeal and a high-quality tailoring feeling, however, from the other one side this also complicate standardizing processes and manufacturing activities. Also, and perhaps more important, since each workers operates in a personal way, trying to implement rapid changeover techniques (i.e. SMED) and mistake-proofing devices (i.e. Poka-Yoka) becomes complicated, since the “autonomation” (i.e. intelligent automation) is not widespread and the processes undergo small changes each time an item changes (Shingo, 1995). A similar reasoning also applies for total production maintenance and total quality management (TQM). Whereas high-level pillars (such as the “Plan Do Check Act” approach, “personnel involvement,” “personal empowerment,” “kaizen,” etc.) can be straightly implemented, technical ones are much more constrained. For instance, the use of generic machines and a lot of manual operations make planned and predictive maintenance rather questionable. Similarly, craftsmanship, product unicity, lack of formal procedures and of standardization pose serious constraints to the use of statistical techniques for processes’ control. 2.3 The framework Based on the above mentioned critical analysis, we decided to use high-level “culture-oriented tools” (i.e. leadership, training, communication, customers’ focus, kaizen events) and low-level “Visual Control” tools (i.e. 5S techniques, single-point lessons, process key performance indicators (KPIs), value stream mapping (VSM) and Poka-Yoka devices) as the main pillars of the framework for lean implementation in luxury-fashion firms. Briefly, the main idea is to use VSM for the preliminary analysis of the “As Is” state and for wastes identification. Next, other visual tools, especially 5S, are used to get rid of (possibly) most of the identified wastes, so as to rapidly achieve a first improvement of the process. Starting from this point, a future state map (i.e. the ideal “To Be” state) is created and used to identify additional and specific lean tools needed to get a further improvement. The main steps of the framework are shown by Table I and are further explained in the following sub-sections. 2.3.1 Step (1): project planning. Once a cross-functional lean team has been formed, the framework starts with the planning of the project, which sets the roadmap of all future activities. During this phase it is fundamental to identify the people (i.e. key users) that will be affected by the change and, next, to involve them in the definition of 993 Lean thinking in the luxury-fashion market

Step Description Objective Tools 1. Project planning The lean project is split into its main phases Responsibilities and deadlines are defined Set the road map for future activities – 1.1 Team formation Starting from the project Statement, a multi-skilled lean team is created Gather all the skills needed to complete the lean project Workers’ skill matrices 1.2 Objective setting Overall goals are defined and/or negotiated with involved people Involve people and force them to reason in terms of concrete objectives SMART technique 1.3 Scheduling Activities are identified and scheduled Milestones and deliveries are fixed Give credibility to the project through the definition of demanding but realistic deadlines and expected outputs Work breakdown structure Design structure matrix Responsibility matrix Gantt chart […] 2. Value stream analysis A value stream is selected and its “As Is” state is mapped Formalize the starting point of the project – 2.1 Value stream’s selection A product family is selected as starting point for the analysis Reduce the complexity of the system to be analyzed Families formation techniques Pareto (ABC) analysis 2.2 Direct observation The production process and the workforce are observed. Information is gathered on the field Get acquainted with the process and gather all the required information Checklist Questionnaires Socio-metric matrices Value stream mapping 2.3 Current state mapping The observed value stream is mapped Formalize the “As Is” state of the value stream Value stream mapping Spaghetti diagrams 2.4 Waste identification From the analysis of the CSM waste are identified and classified Identify the main waste/criticalities to be tackled 7 Wastes Interviews Value stream mapping 3. Waste identification and removal Lean corrective actions are devised, implemented, tested and consolidated Improve the process getting rid of all the identified wastes and criticalities – 3.1 Root cause analysis Identified wastes are analyzed in details and their root causes are identified Facilitate and guide the definition of effective corrective actions Problems’ stratification Ishikawa’s diagrams Process FMEA 3.2 Devising corrective actions Correctives actions are conceptualized to tackle the criticalities emerged at the previous step Improve the value stream through lean initiatives 5S technique Visual management Lean toolkit (if applicable) Focussed groups and lateral thinking 3.3 Prioritization of corrective actions Wastes are ranked in order of criticality Sequence the order of the corrective actions, so as to optimize the use of scarce resources Risk assessment matrix (i.e. severity × occurrence) 3.4 Future state mapping The “To Be” value stream is mapped Formalize the ideal state that should be reached at the end of the project Value stream mapping 4. Implementation and consolidation The solutions are introduced and consolidated KPI are defined and used to assess the quality of the solutions Reach the ideal state and consolidate obtained results Pilots projects Procedures Key performance indicators Table I. The proposed framework 994 IJRDM 43,10/11

SMART (specific, measurable, attainable, realistic and timely) goals. This practice allows formalizing the expected improvements, but it is also permits to get a strong commitment of all the involved people. Setting milestones and deliverables is another important issue, because it gives credibility to the project and forces people to reason in terms of concrete objectives. 2.3.2 Step (2): value stream analysis. The framework proceeds with the direct observation of the (production) process and of the activities performed by the workforce. The aim is: to obtain a precise picture of the current situation; to understand how goods are produced; and to understand which are the most important factors of the manufacturing process. A clear understanding of the “As Is” state will be the starting point to decide future improvements. More specifically, the team starts with the selection of the product, family or product line (i.e. the value stream) that will be analyzed. Next, to get a precise idea of the selected value stream, the lean team should go directly on the shop floor, avoiding relying, exclusively, on the descriptions given by the workforce. Nonetheless, interviewing the workforce may be of help, not only to discover and to clarify eventual divergences, in terms of processes and procedures, among the lean team and the workers, but also to encourage people to bring to light problems and/or criticalities that otherwise would have gone unnoticed. During this phase, in addition to questionnaires and structured interviews, the socio-metric matrices can also be used to capture the informal relationships established among the workforce. This analysis is extremely useful to understand the actual information stream and the presence of contrasts, dislikes and/or other problems. Lastly, once the relevant information has been gathered on the field, the “As Is” situation is (graphically) formalized by drawing the current state map (CSM). This map fixes the starting point of the project and, lately, will be used as the element of comparison to assess future improvements. In order to draw the map we suggest using the VSM icons and the procedure detailed by Braglia et al. (2006); also, in case of complex layout (i.e. job shops), other mapping techniques such as the Spaghetti chart and the material flow diagram (Marangoni et al., 2013) can be used to get a more refined representation of the process under analysis. 2.3.3 Step (3): waste identification and removal. Starting from the CSM, the lean team has to identify the main wastes of the process and conceive possible corrective actions. Doubtless, this operating step is case specific and cannot be generalized. We can say that a correct application of 5S technique and visual control will be the key element for the improvement, but, depending on the system under analysis, other lean tools (such as kanban system, cell design, etc.) could be applied, as further discussed in the Section 3. Also lateral thinking approaches (such as Brainstorming, Delphi method, the 5 Whys method, etc.) can be used to push the generation of new ideas, but wit and past experience remain essential to obtain good solutions. We also note that, in order to properly define effective corrective actions, the analysis should not be limited to the clarification of the effects of the problems, but rather it should be extended to their root causes. Several well-known techniques can be used to this aim, the main ones being the Ishikawa cause and effect diagrams and the process failure mode and effect analysis (Zammori and Gabbrielli, 2012). 2.3.4 Step (4): solutions implementation and consolidation. Before moving to the operational part of the project, the ideal “To Be” state should be (graphically) formalized by drawing the future state map (FSM) of the process. The map should show the required 995 Lean thinking in the luxury-fashion market

corrections actions and should detail the order with which such actions will be performed. It is evident that corrective actions cannot be started at the same time, and so, in order to optimize the use of scares resources, without jeopardizing the prosecution of routine activities, they should be scheduled in a time phased way. To this aim, as one usually does in risk analysis, problems should be ranked in terms of criticality, by considering the probability of occurrence and the severity of their effects. It also advisable to include (on the map) a proper set of KPI that will be used to control the improvements obtained on the shop floor ad to communicate objective results to the top management. To conclude, we note that the FSM can be considered as a sort of project’s baseline and, as such, it may be progressively updated depending on the achieved results. 3. Case studies In order to show the potentialities of the proposed framework, both in production and logistics areas, in the following part of the paper we will describe the main results obtained in a manufacturing facility and in the central warehouse of IFC. 3.1 Lean in a manufacturing facility of leather bags The analyzed facility produces leather bags and employs around 500 workers; leather handbags accounts for more than 95 percent of the overall production, while the remaining capacity is used for processing precious materials and unique accessories. The main features of the facility are summarized in Table II. Specifically, leather arrives pre-cut and production can be divided into three main phases: (1) Preparation phase – the preparation area is configured as a job-shop facility where semi-finished products are manufactured starting from precut leather and reinforcements, which are the main inputs of the process. Due to the high number of set-ups, production is based on large batches of more than 100 units each. In case of extra capacity, semi-finished bags are manufactured also for external companies (i.e. subcontracting). Characteristic Description Procurement process Italian leather is the main raw material, but some bags are made of very valuable skins (i.e. crocodile, ostrich, python, mink, etc.) imported from abroad Production process Cutting and assembly are performed in house, but some other steps of the manufacturing process may be outsourced Retailing Mostly mono-brand boutiques, with few multi-brand boutiques Demand management Make to order (MTO) Sourcing choices Long term partnership and parallel sourcing are used with high-quality Italian suppliers. Multi sourcing is used for less valuable raw materials purchased abroad Manufacturing choices Mostly in-house manufacturing; in case of extra capacity, subcontracting of semi-finished products is allowed Retailing choices In case of under stock and/or of urgent deliveries, priority is given to monobrand boutiques Critical success factors (CSF) Premium quality, exclusivity, made in Italy, craftsmanship, variety Table II. Characteristics of the luxury leather bags’ manufacturer 996 IJRDM 43,10/11

(2) Assembly phase – semi finished products are assembled and the accessories are added to differentiate the product. The assembly area is configured as a mono product assembly line, including coloring, sewing and manual assembly steps. Conversely, the precious handbags are assembled off-line (in dedicated assembly station) by highly skilled and expert workers. (3) Finishing phase – finished product requires a 100 percent quality control. Bags that pass the control are stored in the warehouse, waiting to be shipped worldwide; the others are reworked or discarded. 3.1.1 Case analysis. For the sake of brevity and clarity, in the next part of this section, we will focus only on the main steps of the framework. Details are given below. Value stream identification – the product portfolio is so diversified and products differ so much from each other that production has to be organized in batches. Each time a different kind of handbag is manufactured the (mono) product assembly line is reconfigured and rebalanced in order to process a new product. For these reasons, it was not possible to consider the whole productive process, and the lean team decided to focus attention only on the best-selling hand bag that accounts for the greatest share of the total revenue (i.e. more than 50 percent). Direct observation of the production process and of the workforce – the manufacturing area was visited and snapshots of the critical points were taken, for subsequent analysis. Competences and expertise of the workforce was formalized in a skills matrix that revealed the absence of polyvalent workers. The most critical situation was found in the assembly area, where none of the workers was able to do both sewing and manual assembly, that are the main operations of the manufacturing process. CSM definition – using data gathered at the previous step, the lean team sketched the CSM shown by Figure 1. As can be seen, the manufacturing process is organized as an old fashioned batch and queue production system, with orders pushed from a production area to the next one, without synchronization. Flow is totally absent and there is a high number of stops during the process: upstream and downstream buffers separate each production area and, in addition, the preparation and the assembly departments operate independently, since semi-finished goods are produced well in advance and are stored in the “WIP Warehouse,” where they wait up to a week, before being further processed. Consequently, the firm follows a make-to-order strategy only on the paper, but the facto a make-to-stock approach is used. Indeed, orders from IFC cover a period of over three months and production is scheduled with a traditional material requirement planning system. Thus, we could say that the factory is managed in a push way, as if it was a detached department of IFC. Needless to say that the overall situation was rather critical; as detailed by Table III the lead time is longer than one and a half week (if we also consider the time lost in the initial and ending warehouses the lead time exceeds two weeks), with only a mere 3.1 percent of valuable productive time. Definition of wastes and problems – as above mentioned, problems were mainly due to a pure push-oriented production system. Also, unnecessary movements of the workers and a poorly organized layout were identified as additional causes of poor system’s performances. 997 Lean thinking in the luxury-fashion market

Forecast based on orders received Planned orders (15 days in advance) CUSTOMER 1× Week PRODUCTION CONTROL 2×Week Preparation Area No. workers: 5 Lot: 100units C/T: 17min C/O: [-] Uptime: 82% Av. Time: 7.8h No. workers: 1 Lot: 100 units C/T: 4min C/O: 8min Uptime: 82% Av. Time: 7.8h SUPPLIER No.2 (PET, boxes, wires, colors, accessories) Orders based on needs 1×2 Week 17 minutes 4.5 days 8 minutes 1 day 50 units 20 units 20 units PLANT MANAGER Quarterly orders PREPARATION ASSEMBLY FINISHING SUPPLIER No.1 (skin, reinforcements, accessories, reels, boxes) MRPVisual control and detailed scheduling Inbound Control Warehouse Mng. No. operators: 1+1 C/T: 2min Av. Time: 7.8h 30units to check 60units in stock Coloring Station WIP warehouse No. operators: 1 C/T: 3min Av. Time: 7.8h Line feeding 60 units Assembly Line No. workers: 4 Lot: 20 units C/T: 1.4h C/O: 1h Uptime: 87% Av. Time: 7.8h 20 units Off Line Assembly No. workers: 6 Lot: 20units C/T: 41min C/O: 10min Uptime: 80% Av. Time: 7.8h Self Test 10 units No. workers: 1 Lot: 20units C/T: 6.5min C/O: 2min Uptime: 99% Av. Time: 7.8h 40 units Packaging No. workers: 1 Lot: 100units C/T: 1min C/O: 5min Uptime: 97% Av. Time: 7.8h Weekly schedule updated daily Weekly schedule updated daily Container: 100 units Production Rate: 100 items/week Weekly Shipments on Friday 1.4 hours 3+1 days 6.5 minutes 1days 1 minute 2 days 2.5 days Value Added Time: 2 hours Total Lead Time=8 days Initial and ending warehouses excluded Value Added Time=2 hours Initial and ending warehouses excluded Figure 1. Current state map 998 IJRDM 43,10/11

Fortunately, as quite usual in these situations, the lean team was able to solve a large percentage of the (minor) problems, by means of a straight application of 5S technique, visual control principles and/or creating adequate documentation and operating procedures. For instance, working and transit areas of the shop floor, as well as input and output areas of each workstations where clearly marked on the ground using paints of different colors, single point lessons concerning the proper use of tools and equipment where posted at each workstation, and drawers with shelving were purchased to give perfect visibility of all items/tools and for a faster picking. Further details are given in Table IV. Nonetheless, the number of unsolved problems was still very high and so, in order to simplify the analysis, the lean team decided to group these criticalities into the following five clusters: Layout; methods and equipment; inadequate skills; production system; and inbound logistics. It is worth noting that problems were clustered depending on their root causes and that root causes were discovered by means of Ishikawa’s diagrams. In particular, the choice of this tool was motivated by its simplicity (a full process FMEA would have been too long and time consuming) and by its ability to visualize a problem from many alternative perspectives. In this respect, an interesting example is shown by Figure 2 that is relative to problem No. 3.3. of Table V (i.e. defects of the semi-finished goods). What is interesting to note is that, if Ishikawa’s diagrams were not used, this criticality could have been wrongly associated to the absence of a formal control point; conversely, it turned out that under-skilled workers (unable to assure the desired level of quality) and the absence of formal procedures (for both production and checking) were the real root causes of the problem. Also note how the introduction of a formal control point would have been wasteful and even counterproductive. Indeed it would have permitted to identify defective items but, on the other one side, it would have not reduced the defective rate of No Problem description 5S Visual control Documentation and procedures 1 Absence of procedures, instructions, checklists || | 2 Workstations and buffers are untidy | | 3 Labeling of the equipment is missing or in bad condition | | 4 Identification of components on the shelves is not intuitive | | 5 Working areas and transit zones are not clearly marked | 6 The polishing machine is filthy | 7 Input and output area are not properly marked | 8 Carts are dirty | 9 Semi-finished cannot be traced due to improper documentation | 10 Support material for the bag is not properly arranged | Table IV. Solved problems Critical parameters Value Total lead time (TLT) 8 (days) Value added time (VAT) 2 (hours) Productive time (VAT/TLT) 3.1% Production rate 20 (items/day) Table III. Critical parameters of the CSM 999 Lean thinking in the luxury-fashion market

semi-finished goods produced in the early stages of the process. Also, and perhaps more important, the use a formal control point implies the use of an extra non VA activity and creates an additional point of interruption of the manufacturing flow. Instead, the adopted solution was easier and free of drawbacks, as it was based on easy and cheap techniques such as training plans, supervised training on the field and single-point lessons. As clearly shown by Table V, by operating in this way, the team was able to formulate some general and conceptual solutions and, above all, it was able to rank root causes in order of criticality, depending on their correlation with the negative performance of the process. 3.1.2 Solutions implementation. Problems associated to an improper productive system were ranked as the most critical ones and so, lean implementation at the shop floor started from line balancing. In this respect, a total absence of manufacturing flexibility was certainly the hardest problem to be faced. As above mentioned, bags were made in a mono-product assembly line, which has to be reconfigured at each production change. Therefore, the first idea to reduce change-over time and to boost productivity was to convert the line into a mixed-model assembly line. Yet, production engineers were rather skeptical about this option, as they were sure that productivity would have declined, due to large differences among manufactured bags. In order to verify this assertion, the lean team performed an analysis of process similarities and commonalities, which demonstrated that, although bags differed esthetically, they were quite similar in terms of routings and manufacturing process. Actually, differences among the bags were only due to the accessories (i.e. zips, magnetic snap closures, handles, patches, etc.) and to the color and shape of all the other leather parts. However, these sub-parts arrive pre-cut and they just need to be sewed using a similar manufacturing process (i.e. the routing is always the same, only cycle times and some tools may change from time to time). More specifically, as can be seen from Table VI, which shows the results of the pairwise comparisons among different versions of product Nos 1 and 2, all bags have more than 80 percent of production commonalities. Owing to these results, the mixed-model line was balanced using the Helgeson and Birnie method, generally referred as ranked positional weight technique (Mahto and Kumar, 2012). As known, this common approach works for simple lines and thus, in order DEFECT OF SEMI-FINISHED MATERIAL MACHINERY AND EQUIPMENT PROCESSING METHODS MANPOWER Workers do not follow the existing (non formalized) procedures Absence of formalized procedures Under Skilled Workers A formal control point does not exist Wrong control equipment Figure 2. An example of Ishikawa’s diagram 1000 IJRDM 43,10/11

to readapt it to a mixed-model line, computations were made taking as reference a virtual product obtained considering the maximum cycle time T(i,j),Max for each work variant i at workstation j. In order to further refine the solution, we repeated the computation by substituting T(i,j),Max with T(i,j),Avg, which is the weighted average of the cycle time (of each work variant i at workstation j) based on the production mix. Lastly, starting from these initial solutions line balancing was improved using a greedy technique. No Problem description Cluster Rank Main solution Other solutions 1.1 Frequent movement of workers Layout 5th Cellular layout 1.2 Narrow working environment 1.3 Washing and Coloring are too far away 1.4 Non-optimal position of components Use of flow-racks +KLT containers 1.5 Feeding time are variables Standards feeding routes 2.1 High-waiting time for drying mastic Methods and equipment 3rd Change activities sequences Purchase missing tools 2.2 Drying phase does not signal the end Automate the process 2.3 Complex management of valuable materials 2.4 Batch containers are unstable Use of flow racks 2.5 Identifying small accessories is difficult 5S +visual management 2.6 Sewing takes too much time Buy supporting surfaces 2.7 The zip manufacturing flow is complex 3.1 Low flexibility of the workforce Inadequate skills 2nd Training plan Training on the field Single point lessons Job rotation+job enlargement 3.2 High amounts of chip in the works Autonomous maintenance +5S 3.3 Defects of semi-finished is too high 4.1 WIP accumulates Production system 1st Transition from a push to a hybrid push pull system 4.2 Frequent machine failures Preventive maintenance 4.3 Drying of colors takes a long time Substitute machine 4.4 Reordering policies to feed the line are missing Kanban and supermarket 4.5 The system is non-flexible Use a multi-products line 4.6 OEE fluctuates (performance are not stable) 4.7 LT is too long (i.e. MTO cause delivering delays) 5.1 Suppliers’ deliveries are not synchronized Inbound logistic 4th Redefine logistics contracts Extend distribution requirement planning 5.2 Time is wasted to unpack accessories 5.3 Time is waste to check inbound materials Free pass +suppliers’ certification 5.4 Suppliers use different delivering batches Standardize pallets and containers Table V. Main problems, classification and criticality ranking 1001 Lean thinking in the luxury-fashion market

The obtained solution is based on six manual workstations and one off-line assembly station, designed to perform all the uncommon operations. Obtained results are listed in Table VII. Also, to synchronize material flow, a hybrid push-pull system was used. As shown by the FSM of Figure 3, since the preparation and coloring areas is organized as a job shop with non-linear and complex routings, this part of the process was left unaltered; conversely, all the process downstream of this “decoupling point” are synchronized and pulled by customer’s demand. More specifically, customers’ orders are released at the packaging area, which, acting as pacemaker transmits production orders back to the assembly line. Due to sequence dependent set up times, supermarkets and a dual kanban system are used to synchronize the pacemaker with the assembly line, whereas “off-line assembling” and “self-testing” are directly connected through FIFO lanes. Also note that to reduce setup’s incidence, production kanbans are collected and grouped using the so-called “Set-Up Wheel.” Briefly, its functioning is as follows. The wheel is split in different colored sectors (one for each product variants assembled in the mixed-model line) ordered according to the productive sequence that minimize setups. Any time a production kanban arrives it is added in the corresponding sector of the wheel and only kanbans that are located in the current section of the wheel (clearly indicated by an arrow) can be processed. Anyhow, if the actual sector is empty, the operator can spin the wheel counterclockwise, until he finds a sector with some kanbans. Spinning the wheel counterclockwise ensures the respect of the optimal production sequence. Furthermore, in order to assure flow, a new coloring machine that halves the drying times was purchased and, aiming to improve the so-called “Takt Image” (i.e. to align the workforce to the Takt Time), a kanban blackboard-light was installed on the shop floor to set and visualize the daily production targets. Similarly, aiming to improve internal logistics, 5S techniques were implemented almost in all productive areas. For example, standard boxes and KLT containers (for accessories, semi-finished products, finished products, documents and spare parts) were purchased and all the areas were equipped V1.1 (%) V1.2 (%) V1.3 (%) V2.1 (%) V2.2 (%) V2.3 (%) V1.1 – 95 85 80 80 80 V1.2 95 – 90 80 80 80 V1.3 85 90 – 80 80 80 V2.1 80 80 80 – 95 90 V2.2 80 80 80 95 – 90 V2.3 80 80 80 90 90 – Table VI. Analysis of products’ similarities and commonalities Performance Mono-product line Mixed-model line Number of variants 1 product in a single variant 2 products in six variants Number of stations 4 10 Line efficiency 60% 82% Output rate 20 (units/time) 25 (units/time) Buffers 20 (units) 15 (units) Off-line assembly 40 (min/batch) 12 (min/batch) Table VII. Line balancing results 1002 IJRDM 43,10/11

Forecast based on orders received Planned orders (15 days in advance) CUSTOMER 1×Week PRODUCTION CONTROL 2×Week Preparation Area No. workers: 5 Lot: 100 units C/T: 17min C/O: [-] Uptime: 82% Av. Time: 7.8h No. workers: 1 Lot: 100 units C/T: 4min C/O: 8min Uptime: 82% Av. Time: 7.8h SUPPLIER No.2 (PET, boxes, wires, colors, accessories) Orders based on needs 1×2 Week 17 minutes 2.5 days 8 minutes 6.5 hours 50 units 20 units PLANT MANAGER Quarterly orders PUSH PART PULL PART SUPPLIER No.1 (skin, reinforcements, accessories, reels, boxes) MRPVisual control and detailed scheduling Warehouse Mng. No. operators: 1 C/T: 2min Av. Time: 7.8h 60 units in stock Coloring Station No. operators: 1 C/T: 2min Av. Time: 7.8h Supermarket Line feeding Assembly Line No. workers: 7 Lot: 2 units C/T: 1.2h C/O: 0.8h Uptime: 87% Av. Time: 7.8h Off Line Assembly No. workers: 3 Lot: 20 units C/T: 12min C/O: 10min Uptime: 80% Av. Time: 7.8h Self Test No. workers: 1 Lot: 20 units C/T: 6.5min C/O: 2min Uptime: 99% Av. Time: 7.8h Packaging No. workers: 1 Lot: 100 units C/T: 1min C/O: 5min Uptime: 97% Av. Time: 7.8h Weekly schedule updated daily Weekly schedule Production Kanban Container: 100 units Production Rate: 100 items/week Weekly Shipments on Friday 1.2 hours 14.5 hours 12 minutes 0.6 h 6.5 minutes 0.6 h 2.5 days Value Added Time: 2 hours Total Lead Time=3.7 days Initial and ending warehouses excluded Value Added Time=130 minutes Initial and ending warehouses excluded Change Contract Standardize Free Pass Near Work Stations FIFO FIFO 30 units 15 units 30 units Job Rotation Standard Path 15 units 15 units 5 minutes 0.6 h Set Up Wheel Figure 3. Future state map 1003 Lean thinking in the luxury-fashion market

with proper accessories for cleaning activities. Lastly, as anticipated in the previous sub-section, since the low-quality rate was associated to an under-skilled workforce, the team did not considered statistical quality control and/or other TQM techniques as feasible options. Indeed, supervised training on the field and the use of single-point lessons were considered more than sufficient. 3.1.3 Obtained results. An ongoing assessment of the implemented solutions confirmed both their quality and robustness; even if the vale added time is still low (i.e. 7.1 percent), due to lead time reduction and stabilization the total lead time dropped to 30 hours and due date fulfillment has risen to 75 percent. Other significant results are listed in Table VIII. To be frank, in spite of these interesting results, the performance of the manufacturing plant could be improved further and the ideal target should be a pure JIT production system. To this aim some constraints need to be solved. These are: • Preparation area – this department is organized as a job shop with non-linear and complex routings, a fact that prevents a continuous material flow. Also, the preparation area produces both for the assembly area and for the sub-contractors, another issue that complicates the realization of a pure pull JIT flow. • Finishing area – the main problem of this area is the highly constrained layout and confined working space. Additional problems are: the irregularity of product flow and the absence of an efficient handling system. • Self-test – handbags made by sub-contractors are shipped back for quality control (i.e. self-test). However, delivery plans are not optimized and due dates and quantities are often variables. 3.2 Lean in a central warehouse of luxury products Other interesting activities took place in the logistic hub located in Rome. The hub receives finished goods from the 12 manufacturers facilities owned by IFC (two of which are in north Italy, the other in a district close to Rome) and dispatches replenishment orders of about 16,000 stock keeping units either to the European peripherals warehouses, or directly to Italian mono-brand fashion boutiques. The site occupies a total area of 8,500 m2 , has a turnover of 1,500,000 (items/year) and a maximum capacity of 300,000 items. More specifically, 6,100 m2 are dedicated to storage and this area is subdivided in four distinct modules. This is because some goods have to be stored on shelves (i.e. leather goods, small leather goods, accessories/jewelries), while other ones have to be kept hanged on bars (i.e. belts, ready to wear clothes and furs). 7 muda Reduction (%) Defects and repairs −90 Over-production na Waiting −50 Transport and handling −40 Inventory −50 Over-processing −10 Motion −50 Note: a A make to order policy is used Table VIII. Final results (muda reduction) 1004 IJRDM 43,10/11

Additionally, the hub works both as a central warehouse and as a transit point, in that goods can be received from the manufactures and stored as is to fulfill fashion boutiques’ future demand, or they can be taken to the packaging area for consolidation, before being temporally stored in a dedicated area, waiting to be grouped and dispatched to peripheral warehouses. The first mode of storage is called packaged goods (PG) management , the latter one loose goods (LG) management. Thus, synthetizing, the four modules of the warehouse are the following ones: (1) Storage area for LG, located close to the acceptance area: 4,100 m2 : • module for goods held in shelves: 2,700 m2 ; and • module for hanged goods: 1,400 m2 . (2) Storage area for PG, located close to the shipment area: 2,000 m2 : • module for goods held in shelves: 1,300 m2 ; and • module for hanged goods: 700 m. 3.2.1 Lean solutions for the packaging process of PG goods. With an average demand of 950 (packages/shift), a net available time of seven hours and 45 minutes per shift, the Takt Time of the pack up cycle equals 29 seconds/item. Packaging includes both automated and manual activities. The first ones, used for strapping, weighting and labeling, are not a matter of concern, since machines can complete a package every 24 seconds. Conversely, manual activities detailed by Table IX are the bottleneck of the pack up cycle. As can be seen, given a total time of three minutes and 14 seconds, seven workers are needed to meet the Takt Time. This corresponds to an average workforce’s saturation equal to 81.7 percent, which is decisively lower than that of the automatic machines, equal to 94.2 percent. Almost 41 percent of the total time is due to Activity No. 2, an activity that has been classified as both necessary but non-value added (NNVA) and pure NVA. Specifically, data input is the only necessary task, whereas searching for a proper box type is a totally wasteful operation. Actually, since IFC uses 18 different kinds of boxes, it is common that workers do not have the proper box type at their fingertips whenever needed. So, they have to leave the workstations to gather empty boxes in a dedicated storage area, which is placed at the end of the packaging lines. This displacement is an unnecessary movement of the workforce (i.e. one of the seven main forms of waste) and must be removed. Activity’s description Activity type Duration % 1. Collect: the worker waits for the arrival of the goods NNVA 00:00:21 10.8 2. Prepare: the worker inserts data in the system and gets/ searches the right box NNVA and NVA 00:01:19 40.7 3. Package: the worker places the goods in the box VA 00:00:58 29.9 4. Close: the worker closes the box VA 00:00:25 12.9 5. Cleaning: the worker prepares the station for the next operation NNVA 00:00:08 4.1 6. Repositioning: the worker places extra goods (sent by mistake) in a dedicated area behind the packaging workstation NVA 00:00:03 1.6 Total – 00:03:14 100 Table IX. Manual activities of the packaging process 1005 Lean thinking in the luxury-fashion market

Certainly, to be lean, the optimal solution should be that to reduce the number of boxes types through standardization, but unfortunately this solution cannot be implemented in the short term. Thus, we proposed a sub-optimal, but immediate and efficient solution, by combining two lean principles taken from 5S and from the SMED techniques. These are, respectively, order the work place, and transform internal set-up into external set-up. In accordance to the first principle, each workstation has been equipped with a roll-cage with shelves dedicated to each kind of box (Plate 1). In accordance to the second principle, in order to minimize set up times, at the beginning of a shift each worker goes to the storage area and replenishes the roll-cage with missing boxes. This additional task takes about ten minutes and so the net available time drops from of seven hours and 45 minutes to seven hours and 35 minutes per shift. Nonetheless, this loss of time is widely balanced by the simplification of Activity No. 2 (i.e. boxes are always available), which now takes only 35 seconds to be completed. It is worth noting that, thank to this reduction, six workers instead of seven are sufficient to meet the desired Takt time of 29 seconds/item. Also, in this configuration, the average workforce’s saturation has increased to 87 percent, a value that is much more aligned with that of the packaging machines. Plate 1. The adopted roll cage 1006 IJRDM 43,10/11

Lastly, to focus people on the Takt time, an electronic Andon Board was installed to show in real time the following KPIs: • total number of packaged goods; • total elapsed time; • average time per packaging; • average composition of the packages (in terms of number and kind of contained goods); and • variation with respect to the standard time and to the average performance evaluated over the last week. Accordingly to the visual control concept, these KPIs, evaluated at an aggregated level on a weekly basis, are printed, posted and discussed by lean circles, so as to highlight and tackle potential problems. 3.2.2 Lean solutions for picking. With regard to picking activities, an interesting intervention concerned the storage of leather belts. As shown by Plate 2, to avoid wrinkling or other damages, these items must be hung to bars (as in a wardrobe), a condition that makes their identification and picking a non-trivial task. Indeed, although the information systems automatically prints a picking list that specifies the picking routes and the location of the stock keeping units (SKUs), a high share of the average picking time is lost (i.e. two minutes on 13) to find the correct SKU. This is mainly due to the following reasons: the label that identifies the SKU is placed randomly and it is not always easy to be found; different kind of SKU (in terms of both products kind and size) may be hung on the same bar; and belts are kept inside a PET wrapping that does not allow a visual recognition of the good. Plate 2. Storage of leather belts 1007 Lean thinking in the luxury-fashion market

Also in this case, an appropriate use of 5S and visual control principles led to the redefinition of the stocking criteria and to the introduction of smart tags (see Plate 3) that facilitate a visual identification of the required good. More specifically, the following solutions were implemented: • random allocation was replaced by a dedicated allocation, with each bar corresponding to a specific kind of belt; • bars were subdivided into four sub-section, one for each possible size of the leather belts; • larger size were associated to high-levels bar, so as to minimize probability of damages during picking activities; and • smart tags were used to visually identify the location of SKUs. Thanks to these modifications, with a negligible economic investment, the average picking time dropped to ten minutes and 30 seconds, a value that corresponds to an increase of efficiency of about 19.2 percent. 4. Conclusions and future developments The paper pointed out that lean has emerged as new managerial paradigm capable to boost operational performance and to fulfill customers’ needs in a more thorough way. However, the paper also noted that declining these paradigms into operating processes remains a challenging task. A universal framework does not exist and lean principles must be re-interpreted, dependently on the system under analysis. This drawback seems to be particularly important in the luxury-fashion market, since the specificities of this sector greatly differ from those of high-volume low-variety manufacturers, where lean was originated. Indeed, although the capability to overcome tradeoffs among costs, quality and time has become a must in this business too, applications in the luxury-fashion market are still rare. This is because, features as demand variability and seasonality, wide product’s range, product customization and workmen craftsmanship are generally considered as insurmountable barriers for lean. Plate 3. Smart tags 1008 IJRDM 43,10/11

Consequently, most of the times, supply chain optimization is indicated as the only possible solution and lean interventions at the shop floor level are not even considered as feasible options. Moving from these premises, in order to see is these barriers are really insurmountable, an empirical approach was followed. Specifically, a comprehensive framework for lean implementation was developed and fully implemented using the experience gained on the field, during an extensive lean project supported by IFC, an important Italian brand that operates worldwide in the fashion-luxury market. The project, coordinated by a team of academicals and top managers, covered IFC’s logistic network and 12 of its manufacturing facilities. Aiming to optimize operations, to rationalize flows and to increase flexibility both in manufacturing and logistics, many critical issues were found and several solutions were devised and implemented as pilot projects, with outstanding preliminary results. Due to space constraints, it was not possible to give full details of the whole project and so the paper focusses exclusively on some of the most interesting activities. Still, similar solutions and similar results were obtained in almost all the other investigated processes. Without going into details, what emerged from the analysis is the fact that, as for most of the businesses where the focus on operational costs has been traditionally considered as a marginal issue, almost all the operations that we analyzed were “old fashioned” and offered fertile ground for improvement. Also, and perhaps more important, we noticed that, many times, barriers to lean do not really exist, but they are only preconceptions due to an old fashioned mindset of the managers that have worked for long in the luxury-fashion market. For instance, making reference to the above mentioned case study, when the lean team suggested converting a mono-product line into a mixed-model one, production engineers were rather skeptical, for they were sure that productivity would have declined, due to the existing differences among manufactured handbags. However, the fact that, due to different materials and components, each SKU must have a specific Bill of Material and must be considered as a specific code for the information system, does not imply that they cannot be aggregated into families, similar in terms of routings and/or manufacturing process. De facto, in the present application, variability was lower than expected and it was possible to design and balance work cells capable to accommodate manufacturing differences, among the manufactured handbags. Obviously, since results are case specific, trying to infer general considerations may be hazardous. Nonetheless, due to the relevant dimension of the project and considering the importance of the investigated company, we think that these results could be considered more than a clue concerning the real potentialities of lean in the luxury-fashion market. Lastly, we note that, although our framework can be helpful for practitioners, the success of a lean project largely depends on the wit and on the expertise of the lean team and, above all, on the commitment of the top management and of all involved people. Without these features, a lean project will never succeed. References Achanga, P., Shehab, E., Roy, R. and Nelder, G. (2006), “Critical success factors for lean implementation within SMEs”, Journal of Manufacturing Technology Management, Vol. 17 No. 4, pp. 460-471. Alvarez, R., Calvo, R., Pena, M.M. and Domingo, R. (2009), “Redesign an assembly line through lean manufacturing tools”, International Journal of Advanced Manufacturing Technology, Vol. 43 No. 9, pp. 949-958. 1009 Lean thinking in the luxury-fashion market

Arienti, P. (2013), “Global power of luxury goods 2014”, paper presented at the 6th Luxury Summit, Milan, June 14. Arlbjørn, J.S., Freytag, P.V. and De Haas, H. (2011), “Service supply chain management: a survey of lean application in the municipal sector”, International Journal of Physical Distribution & Logistics, Vol. 41 No. 3, pp. 277-295. Aziz, R.F. and Hafez, S.M. (2013), “Applying lean thinking in construction and performance improvement”, Alexandria Engineering Journal, Vol. 52 No. 4, pp. 679-695. Bertolini, M., Braglia, M., Romagnoli, G. and Zammori, F. (2013), “Extending value stream mapping: the synchro MRP case”, International Journal of Production Research, Vol. 51 No. 18, pp. 5499-5519. Bhamu, J., Khandelwal, A. and Sangwan, K.S. (2013), “Lean manufacturing implementation in an automated production line: a case study”, International Journal of Services and Operations Management, Vol. 15 No. 4, pp. 411-429. Bhardwaj, V. and Fairhurst (2010), “Fast fashion: response to changes in the fashion industry”, The International Review or Retail. Distribution and Consumer Research, Vol. 20 No. 1, pp. 165-1173. Bonaccorsi, A., Carmignani, G. and Zammori, F. (2011), “Service value stream management (svsm): developing lean thinking in the service industry”, Journal of Service Science and Management, Vol. 4 No. 4, pp. 428-439. Boyle, T.A. and Scherrer-Rathje, M. (2009), “An empirical examination of the best practices to ensure manufacturing flexibility: lean alignment”, Journal of Manufacturing Technology Management, Vol. 20 No. 3, pp. 348-366. Braglia, M., Carmignani, G. and Zammri, F. (2006), “A new value stream mapping approach for complex production systems”, International Journal of Production Research, Vol. 44 Nos 18-19, pp. 3929-3952. Braglia, M., Frosolini, M. and Zammori, F. (2009), “Uncertainty in value stream mapping analysis”, International Journal of Logistics Research and Applications, Vol. 12 No. 6, pp. 435-453. Bruce, M., Daly, L. and Towers, N. (2004), “Lean or Agile: a solution for supply chain management in the clothing and textile industry?”, International Journal of Operations & Production Management, Vol. 24 No. 2, pp. 151-170. Brun, A., Caniato, F., Caridi, M., Castelli, C., Miragliotta, G., Ronchi, S., Sianesi, A. and Spina, G. (2008), “Logistics and supply chain management in luxury fashion retail: empirical investigation of italian firms”, International Journal of Production Economics, Vol. 114 No. 2, pp. 554-570. Caniato, F., Caridi, M., Castelli, C. and Golini, R. (2009), “A contingency approach for SC strategy in the Italian luxury industry: do consolidated models fit?”, International Journal of Production Economics, Vol. 120 No. 1, pp. 176-189. Caniato, F., Caridi, M., Castelli, C. and Golini, R. (2011), “Supply chain management in the luxury industry: a first classification of companies and their strategies”, International Journal of Production Economics, Vol. 133 No. 2, pp. 622-633. Chen, L. and Meng, B. (2010), “The application of value stream mapping based lean production system”, International Journal of Business and Management, Vol. 5 No. 6, pp. 203-209. Christopher, M. and Peck, H. (1997), “Managing logistics in fashion markets”, The International Journal of Logistics Management, Vol. 8 No. 2, pp. 63-74. D’Arpizio, C. (2013), “Luxury goods worldwide market study”, Bain & Company’s report, October 28, available at: www.bain.com/publications/articles/luxury-goods-worldwide-marketstudy-fall-2013.aspx 1010 IJRDM 43,10/11

D’arpizio, C. (2014), “Bain & Company’s 2014 annual global luxury study”, Bain & Company’s report, October 14, available at: www.bain.com/about/press/press-releases/bain-and-companys-2014- annual-global-luxury-study.aspx Demeter, K. and Matyusz, Z. (2011), “The impact of lean practices on inventory turnover”, International Journal of Production Economics, Vol. 133 No. 1, pp. 154-163. Green, K.W., Inman, R.A., Birou, L.M. and Whitten, D. (2014), “Total JIT (T-JIT) and its impact on supply chain competency and organizational performance”, International Journal of Production Economics, Vol. 147 Part A, pp. 125-135. Hines, P., Holweg, M. and Rich, N. (2004), “Learning to evolve: a review of contemporary lean thinking”, International Journal of Operation and Production Management, Vol. 24 No. 10, pp. 994-1011. Hirano, H. (1995), 5 Pillars of the Visual Workplace, Productivity Press, Portland, OR. Hodge, G.L., Goforth, R.K., Joines, J.A. and Thoney, K. (2011), “Adapting lean manufacturing principles to the textile industry”, Production Planning & Control, Vol. 22 No. 3, pp. 237-247. Holweg, M. (2007), “The genealogy of lean production”, Journal of Operations Management, Vol. 25 No. 2, pp. 420-437. Hüttmeir, A., De Treville, S., Van Ackere, A., Monnier, L. and Prenninger, J. (2009), “Trading off between heijunka and just-in-sequence”, International Journal of Production Economics, Vol. 118 No. 2, pp. 501-507. Kapferer, J.N. and Bastien, V. (2009), “The specificity of luxury management: turning marketing upside down”, Journal of Brand Management, Vol. 16 No. 5, pp. 311-322. Kuldip, J.B. and Sangwan, S. (2014), “Lean manufacturing: literature review and research issues”, International Journal of Operations and Production Management, Vol. 3 No. 7, pp. 470-876. Lewis, M. (2000), “Lean production and sustainable competitive advantage”, International Journal of Operations & Production Management, Vol. 20 No. 8, pp. 959-978. Li, H.H.J. and Tan, K.H. (2004), “SMEs business growth model: a medium to big effort”, International Journal of Management and Enterprise Development, Vol. 1 No. 3, pp. 195-207. Mahto, D. and Kumar, A. (2012), “An empirical investigation of assembly line balancing techniques and optimized implementation approach for efficiency improvements”, Global Journal of Researches in Engineering, Vol. 2 No. 3, pp. 1-14. Marangoni, G., Romagnoli, G. and Zammori, F. (2013), “Multiple value stream mapping: how to implement work load control in complex systems”, International Conference on Industrial Engineering and Systems Management (IESM), Rabat, October 28-30. Masson, R., Iosif, L., MacKerron, G. and Fernie, J. (2007), “Managing complexity in agile global fashion industry supply chains”, The International Journal of Logistics Management, Vol. 18 No. 2, pp. 238-254. Melton, T. (2005), “The benefits of lean manufacturing: what lean thinking has to offer the process industries”, Chemical Engineering Research and Design, Vol. 83 No. 6, pp. 662-673. Muda, S. and Hendry, L. (2003), “The SHEN model for MTO SMEs: a performance improvement tool”, International Journal of Operations & Production Management, Vol. 23 No. 5, pp. 470-486. Nugroho, I.A., Riastuti, U.H. and Iridiastadi, H. (2012), “Performance improvement suggestions for ground handling using lean solutions approach”, International Congress on Interdisciplinary Business and Social Science, ICIBSoS, Jakarta, December 1-2. Olivera, N., Schabb, L. and Holwegb, M. (2007), “Lean principles and premium brands: conflict or complement?”, International Journal of Production Research, Vol. 45 No. 16, pp. 3723-3739. 1011 Lean thinking in the luxury-fashion market