Expressive Commerce and Its Application to Sourcing

Sourcing professionals buy several trillion dollars worth of goods and services yearly. We introduced a new paradigm called expressive commerce and applied it to sourcing. It combines the advantages of highly expressive human negotiation with the advantages of electronic reverse auctions. The idea is that supply and demand are expressed in drastically greater detail than in traditional electronic auctions, and are algorithmically cleared. This creates a Pareto efficiency improvement in the allocation (a win-win between the buyer and the sellers) but the market clearing problem is a highly complex combinatorial optimization problem. We developed the world’s fastest tree search algorithms for solving it. We have hosted $19 billion of sourcing using the technology, and created $2.1 billion of hard-dollar savings. The suppliers also benefited by being able to express production efficiencies and creativity, and through exposure problem removal. Supply networks were redesigned, with quantitative understanding of the tradeoffs, and implemented in weeks instead of months. Historical Backdrop on Sourcing Sourcing, the process by which companies acquire goods and services for their operations, entails a complex interaction of prices, preferences, constraints, and many non-price attributes. The buyer's problem is to decide how to allocate the business across the suppliers. Traditionally, sourcing decisions have been made via manual in-person negotiations. The advantage is that there is a very expressive language for finding, and agreeing to, win-win solutions between the supplier and the buyer. The solutions are implementable because operational constraints can be expressed and taken into account. On the downside, the process is slow, unstructured, and nontransparent. Furthermore, sequentially negotiating with the suppliers is difficult and leads to suboptimal decisions. (This is because what the buyer should agree to with a supplier depends on what other suppliers would have been willing to agree to in later negotiations.) The 1-to-1 nature of the process also curtails competition. These problems have been exacerbated by a dramatic shift from plant-based sourcing to global corporate-wide Copyright © 2006, American Association for Artificial Intelligence (www.aaai.org). All rights reserved. (category-based rather than plant-based) sourcing since the mid-1990’s. This transition is motivated by a corporation’s desire to leverage its spend across plants in order to get better pricing and better understanding and control of the supply chain while at the same time improving supplier relationships. (See, e.g., (Smock 2004).) This transition has yielded significantly larger sourcing events that are inherently more complex. During this transition, there has also been a shift to electronic sourcing where suppliers submit offers electronically to the buyer. The buyer then decides, using the software, how to allocate the business. Advantages of this approach include speed of the process, structure and transparency, global competition, and simultaneous negotiation with all suppliers (which removes the difficulties associated with the speculation about later stages of the negotiation process, discussed above). The most famous class of electronic sourcing systems which became popular in the mid-1990s through vendors such as FreeMarkets (now Ariba), Frictionless Commerce, and Procuri is a reverse auction. The buyer groups the items into lots in advance, and conducts an electronic descending-price auction for each lot. The lowest bidder wins. (In some cases “lowness” is not measured in terms of price, but in terms of an ad hoc score which is a weighted function that takes into account the price and some non-price attributes such as delivery time and reputation.) Reverse auctions are not economically efficient, that is, they do not generally yield good allocation decisions. This is because the optimal bundling of the items depends on the suppliers' preferences (which arise, among other considerations, from the set, type, and time-varying state of their production resources), which the buyer does not know at the time of lotting. Lotting by the buyer also hinders the ability of small suppliers to compete. Furthermore, reverse auctions do not support side constraints, yielding two drastic deficiencies: 1) the buyer cannot express her business rules; thus the allocation of the auction is unimplementable and the “screen savings” of the auction do not materialize in reality, and 2) the suppliers cannot express their production efficiencies (or differentiation), and are exposed to bidding risks. In short, reverse auctions assume away the complexity that is inherent in the problem, and dumb down the events rather than embracing the complexity and viewing it as a driver of opportunity. It is therefore not surprising that there are strong broad-based signs that reverse auctions have fallen in disfavor. The New Paradigm: Expressive Commerce In 1997 it dawned on me that it is possible to achieve the advantages of both manual negotiation and electronic auctions while avoiding the disadvantages. The idea is to allow supply and demand to be expressed in drastically more detail (as in manual negotiation) while conducting the events in a structured electronic marketplace where the supply and demand are algorithmically matched (as in reverse auctions). The new paradigm, which we call expressive commerce (or expressive competition), was so promising that I decided to found CombineNet, Inc. to commercialize it. The finer-grained matching of supply and demand yields Pareto improvements (i.e., win-win solutions) between the buyer and the suppliers. However, matching the drastically more detailed supply and demand is an extremely complex combinatorial optimization problem. We developed the world’s fastest algorithms for optimally solving it. These algorithms are incorporated into the market-clearing engine at the core of our flagship product, the Advanced Sourcing Application Platform (ASAP). Expressive commerce has two sides: expressive bidding and expressive allocation evaluation (also called expressive bid taking) (Sandholm and Suri 2001).