AI for fashion supply chain

The fifth international workshop on fashion and KDD

24 August 2020, San Diego, California - USA

-

1:00PM - 5:00PM Pacific Time

Background

Fashion is a multi-billion-dollar industry with social and economic implications worldwide. One of the biggest problems confounding several fashion houses is the problem of unsold dead inventory. Despite having vast amounts of historical data pertaining to product, sales and inventory, only about 50-60% of the products sell well and rest go through severe mark downs. This is primarily due to volatile nature of fashion trends and fashion brands producing several new products to catch up with the trend. Fashion, i.e., “prevailing style during a particular time”, by design and definition changes by season, and so any unsold inventory at the end of each season is generally liquidated. While smaller designers and retailers generally move the merchandise to discount stores, some top design houses resort to recycling or even destroying the merchandise to avoid dilution of their brand. The environmental impact of the industry is even worse with the US EPA estimating 16 million tons of textile waste generated in 2016 in just the US.

Behind many fashion brands is a highly complex supply chain. Unsold merchandise/inventory is mainly due to mismatch between supply and demand. It could be that the inventory has been over-produced or not distributed properly at the right location and at the right time, mainly due to inaccurate demand forecastsand inefficiencies in supply chain planning. With the advent of modern AI technologies and vast amounts of (structured and unstructured) fashion data the impact on the fashion industry could be transformational and eventually enable intelligent self-correcting sustainable supply chains so that excess inventory is minimized.

The stakes are high for fast-fashion retailers and the insights provided by data can help facilitate earlier trend detection and more accurate demand forecasts to help build a more flexible and faster supply chain that reacts to market trends, manage assortment and inventory tuned to the local tastes and set prices and markdown optimally.

The goal of this workshop is to gather people from academia, industry, and startups working at the intersection of fashion, AI, data mining and knowledge discovery, and supply chain to further the technology and its adoption. The first, second, third and fourth international workshops on fashion and KDD were organized at KDD 2016 in San Francisco, KDD 2017 in Halifax, KDD2018 in London and KDD 220 in Anchorage respectively, and engendered scholarly debate among experts and led to several innovative ideas.

This year, responding to attendee feedback and increasing focus on responsible and sustainable supply chains, we are pivoting the workshop more towards the entire fashion supply/value chain rather than just the consumer facing use-cases. Indeed, in January 2020, the World Economic Forum Annual Meeting in Davos, Switzerland also recognized the growing importance of building more sustainable supply chains.

Topics of Interest

We hope this workshop will bring together all the researchers, practitioners, and interested audiences to explore the open problems, applications, and future directions in this field. We believe that the fashion industry introduces several interesting learning problems that are either not studied or scarcely studied in the past and can attract great interest in the general KDD community given their practical implications. Suggested topics include but are not limited to

  • AI for fashion supply/value chain
  • Deep learning for fashion
  • Sales and demand forecasting
  • Improved forecasting incorporating external events like weather, events etc.
  • Attribute based demand forecasting
  • New product demand forecasting
  • Spatial-temporal hyper-local demand forecasting 
  • New stores sales forecasting
  • Demand transference models 
  • Discrete choice models
  • Trend analysis and forecasting
  • Sentiment analysis
  • Hyper-local assortment planning
  • Inventory allocation 
  • Store clustering
  • Inventory policies  and dynamic auto-replenishment
  • Markdown optimization 
  • Stock re-allocation and inter store transfer 
  • Algorithmic product design 
  • Omni-channel order fulfillment and inventory planning
  • Visual search for fashion e-commerce
  • Fashion image understanding and auto-tagging of apparel
  • Virtual personal fashion assistants
  • Recommendation engines for fashion
  • AI tools for fashion designers, buyers, merchandisers, and consumers
  • Sustainable supply chains
  • Improving provenance and trust in the fashion supply chain
  • Circular economy and methods and policies to encourage reuse and recycling of materials
  • Handling disruptions in fashion supply chain 
  • Ideas from other industries with rapid obsolescence

Workshop Schedule


24th August, 2020 ( 01:00PM - 5:00PM ) Pacific time
Local timings
24th August, 2020 ( 04:00PM - 8:00PM ) New York Time
25th August, 2020 ( 01:30AM - 5:30AM ) Indian Standard Time

1:00 - 1:15 pm - Introduction
    Welcome and Invited Talk by Vikas Raykar on AI for fashion supply/value chain

1:15 - 2:00 pm - Invited Talk
    Invited Talk by Chaithanya Bandi on Personalization via Active Learning - Tractable algorithms with performance guarantees

2:00 pm - 3:00 pm - Oral Paper Presentations (Session 1)
  • 2:00 - 2:15 pm - AI Assisted Apparel Design.
  • 2:15 - 2:30 pm - Explainable AI based interventions for pre-season decision making in fashion retail.
  • 2:30 - 2:45 pm - Breaking Moravec's Paradox: Visual-Based Distribution in Smart Fashion Retail.
  • 2:45 - 3:00 pm - Hyper-local sustainable assortment planning.

03:00 - 03:15 pm - Break
03:15 pm - 05:00 pm - Oral Paper Presentations (Session 2)
  • 03:15 - 03:30 pm - An Application of Newsboy Problem in Supply Chain Optimisation of Online Fashion E-Commerce.
  • 03:30 - 03:45 pm - Product age based demand forecast model for fashion retail.
  • 03:45 - 04:00 pm - Intelligent Warehouse Allocator for Optimal Regional Utilization.
  • 04:00 - 04:15 pm - Price Optimization in Fashion E-commerce.
  • 04:15 - 04:30 pm - Deep Contextual Embeddings for Address Classification in E-commerce.

Invited Speakers

Chaithanya Bandi
Associate Professor of Operations, Kellogg School of Management
Invited talk on Personalization via Active Learning - Tractable algorithms with performance guarantees

Chaithanya Bandi is an Assoicate Professor of Operations at the Kellogg School of Management. He received a Ph.D in Operations Research from MIT in 2013, and a Bachelors of Technology in Computer Science and Engineering from IIT Madras in 2008. He has worked and collaborated with various Technology companies including Google, Amazon, Facebook, IBM and finance companies including Blackrock, Goldman Sachs, Lehman Brothers, Investment Technology group etc.

Chaithanya is broadly interested in the problems of decision making under uncertainty, incomplete information and risk with applications to operations management. In particular, I have focussed on developing Robust Optimization based models to formulate key problems in applications such as queueing control, risk optimization, mechanism design, and online algorithms; with applications ranging from e-commerce, healthcare, crowdsourcing, data-centers, and cloud-computing.


Accepted Papers



Organizers

Advisory Panel

Program Committee Members

Submission Guidelines

We solicit submission of papers of papers of 4 to 10 pages representing reports of original research, preliminary research results, case studies, proposals for new work and position papers. We also seek poster submissions based on recently published work (please indicate the conference published).

All papers will be peer reviewed, single blind (i.e. author names and affiliations should be listed). If accepted, at least one of the authors must attend the workshop to present the work. The submitted papers must be written in English and formatted in the double column standard according to the ACM Proceedings Template, Tighter Alternate style. The papers should be in PDF format and submitted via the EasyChair submission site. The workshop website will archive the published papers.

For more information or any clarifications please email aiforsupplychain.kdd@gmail.com

  • Paper Submission Deadline: June 1, 2020 June 8, 2020
  • Acceptance Notifications: June 15, 2020 June 23, 2020
  • Workshop date: August 24, 2020

All deadlines are at 11:59 PM Pacific Standard Time.

Submit