The Internet has led to the development of online markets, and computer scientists have designed various auction algorithms, as well as automated exchanges for standardized commodities; however, they have done little work on exchanges for complex non-standard goods. We present an exchange system for trading complex goods, such as used cars or non-standard financial securities. The system allows traders to represent their buy and sell orders by multiple attributes; for example, a car buyer can specify a model, options, colour, and other desirable features. Traders can also provide complex price constraints, along with preferences among acceptable trades; for instance, a car buyer can specify dependency of an acceptable price on the model, year of production, and mileage. We describe the representation and indexing of orders, and give algorithms for fast identification of matches between buy and sell orders. The system identifies the most preferable matches, which maximize trader satisfaction, and it allows control over the trade-off between speed and optimality of matching. It supports markets with up to 300 000 orders, and processes hundreds of new orders per second.

We present an automated exchange for trading complex goods, such as used cars, which allows traders to describe desirable purchases and sales by multiple attributes. The developed exchange system supports markets with up to 300,000 orders.

We present an exchange system for trading nonstandard goods, such as used cars. We explain the representation of desirable purchases and sales, describe data structures for fast identification of matches between buyers and sellers, and give experiments on the system's performance.