Research Statement
The IBM Deep eCommerce Department develops techniques in the areas of decision analysis, combinatorial optimization,
machine learning and statistics, and applies them to real-world problems in the area of electronic sourcing and supply
chain management. In this context, we have been looking at problems such as:
- Winner determination
- Preference elicitation
- Bid price estimation
- Computational mechanism design
- Automated mapping of commodity codes
Winner determination in auctions has been a central problem in our work.
Traditional auction mechanisms allow price-only negotiations for which the winner
determination is a computationally simple task. However, the need for new auction
mechanisms that allow complex bids such as bundle bids and multi-attribute bids has
been raised in many situations. The winner determination in these auctions is a
computationally hard problem. The computational complexity has been a significant
hurdle for the widespread use of these advanced auction models. Examples are:
- Multi-attribute auctions
- Combinatorial auctions
- Multi-unit auctions
- Volume discount auctions
- Evaluation of configurable offers
| 
Other
projects in
