Simulating Reverse Auction Bidding of Construction Contracts Dr. Khaled Nassar University of Sharjah Sharjah, UAE Simulation using agents has gained a wide prominence in a wide area of applications. In particular agent-based negotiation is concerned with simulating the behavior of parties interacting with each other to reach an agreement. In this paper, a construction-specific model for negotiation between agents representing contractors in a Reverse Auction Bid (RAB) is presented. RAB is a fairly new electronic bidding model where contractors bid on a particular contract by iteratively lowering their bids. We specifically present a model that can deal with many-parties, single issue, multiple- encounter negotiations. A computerized system is developed to test the model. An example of the system is presented for clarification. Keywords: Teaching Evaluation, Student Course Evaluation, Relational Databases Autonomous agents, construction contracts, reverse auction bidding, negotiation Introduction Negotiation is a central characteristic of the construction process. There are a number of different negotiation theories and techniques that can be used in a variety of situations (Barcharach and Newnham 2000, Bell 2001, Bacharach and Lawler1981, Anumba and Newnham 2000, Ren et al 2003). In particular, negotiation involving construction contracts are usually of a type called service-oriented negotiations (Zeuthen 1975). Service-oriented negotiations are a type of negotiation where the purpose is to reach an agreement about the provision of a service by one side for another. Nevertheless, all types of negotiations entail two separate but interconnected elements: a rational decision making element and a psychological element. In order to improve the outcomes of negotiations, automated negotiation using autonomous agents have recently become one of the fundamental decision models studied. The main perceived benefit of using agent-based negotiation technology is the removal of the emotional component of the negotiations that so often result in less than Pareto-optimal agreements. A Pareto efficient outcome is one in which there is no other agreement that would result in both parties being better off. If there is an outcome that would have made both better off, the decision reached is not Pareto-optimal. Pareto-optimal agreements usually are attainable when there are options to be negotiated in the terms of the agreement. In order to investigate the properties of RAB, an autonomous bidding model is developed in this paper. The model aims at investigating basic properties of the RAB model. The remainder of this paper is organized as follows; first, an introduction to negotiation is offered. This is followed by a discussion of Reverse Auction Bidding as a new method of construction bidding and differences between RAB and traditional bidding are outlined. Next, the developed model is presented and example simulations of autonomous bidding are described. Finally, conclusions are drawn and recommendations for future work are presented.