1 / 30

Gürdal Ertek, Merve Ayşe Can, Firdevs Ulus

Benckmarking the Turkish Apparel Industry Through Data Envelopment Analysis (DEA) and Data Visualization. Gürdal Ertek, Merve Ayşe Can, Firdevs Ulus. Retailing.

jeff
Download Presentation

Gürdal Ertek, Merve Ayşe Can, Firdevs Ulus

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Benckmarking the Turkish Apparel IndustryThrough Data Envelopment Analysis (DEA) andData Visualization Gürdal Ertek, Merve Ayşe Can, Firdevs Ulus

  2. Retailing • “Retailing consists of the sale of goods/merchandise for personal or household consumption either from a fixed location or from a fixed location and related subordinated services.”

  3. Our Study • Benchmarking the Turkish apparel retail industry • 39 companies included • Financial and operational data used together • Methodologies • Data Envelopment Analysis (DEA) • Data visualization • Applied together earlier (ex: Ulus et al., 2006) • Extended in this study with tile visualization

  4. MethodologiesData Envelopment Analysis (DEA) • Methodology for comparing efficiency of a group of entities • Decision Making Units (DMUs) • Nonparametric • Based on optimization modeling • Few assumptions about the units and magnitudes of data • Nonnegative data

  5. MethodologiesData Envelopment Analysis (DEA) • Uses data regarding the DMUs • Inputs (Resources consumed) • Outputs (Desired outcomes produced) • Represents the efficiencies of these entities as a single computed value ranging from 0 to 1 • Efficiency scoresof the DMUs • Also returns the reference sets • Efficient frontieris the boundry drawn by the efficient DMUs

  6. MethodologiesData Envelopment Analysis (DEA) • Trivial Case I • Single input X and single output Y • Sort the DMUs wrt decreasing efficiencies: Y/X • Normalize the efficiency values to obtain efficiency scores

  7. MethodologiesData Envelopment Analysis (DEA) • Trivial Case II • Inputs X1 & X2 and Single output Y

  8. MethodologiesData Envelopment Analysis (DEA) • Trivial Case III • Single input X and Outputs Y1 & Y2

  9. MethodologiesData Envelopment Analysis (DEA) • Non-Trivial Case • Multiple Inputs and Outputs

  10. MethodologiesData Visualization • Visually understanding data • Detecting important patterns • Coming up with hypotheses • Emergence of Information Visualization • Data mining, computer graphics, human-computer interaction (HCI), explanatory data analysis • New styles of visualizations • Spence (2001)

  11. Data • Turkishtime, August 2006 • Joint effort with AMPD • Missing data tediously collected through telephone and e-mail contact • Data extended with new fields • Foundation year • Product types • Genders served

  12. Data • 39 retailers included in the DEA model • Inputs • Number of stores • Number of corners* • Total sales area (m²) • Number of employees • Output • Annual sales revenue

  13. Model • BCC input oriented DEA model • DEA carried out through DEA-Solver software • Developed by Kaoru Tone • Included in the DEA book by Cooper et al. (2006) • Efficiency scores obtained and visually analyzed • Original fields • Derived fields • Other relevant data

  14. Analysis Omniscope www.visokio.com Miner3D www.miner3d.com

  15. Efficiency vs. Foundation Year

  16. Efficiency wrtEmployee per Area & Area per Store

  17. Darker colors denote higher efficiency scores.

  18. Efficiency vs. No of Stores

  19. Larger sizes denote larger area/store.

  20. Efficiency wrt Product Types C H K U S

  21. C: Classic S: Sports A: At least three types H: Shoes and accessories K: Kids U: Underwear

  22. C: Classic S: Sports A: At least three types H: Shoes and accessories K: Kids U: Underwear Darker colors denote higher efficiency scores.

  23. Efficiency wrt Gender M W K

  24. M: Men’s W: Women’s K: Kids’ Darker colors denote higher efficiency scores.

  25. Conclusions • Efficiency independent of foundation year • Higher variability in efficiency scores of younger companies • A group of efficient companies showing a peculiar pattern • Employee/Area increases linearly with Area/Store

  26. Conclusions • Number of stores is not a determinant of efficiency • Except LC WAIK and KOTON • TWEEN performing badly despite large no of stores and large sales area • Low efficiency of kids’ apparel retailers • Possibly due to competition with global brands such as Chicco (not included in the study) • None of the companies have efficiency score within the range [0.75, 0.99]

  27. Conclusions • Harsh competition in classic wear • Too many companies operating • Efficiency scores < 0.63 (except GERMIR) • Also fierce competition with companies in group A (BOYNER, YKM, KOTON, DESIGN) • Typical characteristics of efficient companies • Large sales area • Retailing broad range of products • Serving both genders

  28. Future Research • Statistically testing the proposed hypotheses • Visualizing the reference sets • Searching the literature and comparing results • Incorporating other data fields • Considering other retailing subsectors and retail sectors in other countries

  29. Thank you…

More Related