E-procurement has become an important function of enterprise information systems. The process of e-procurement includes the automatic definition of product requirements, search and selection for suppliers, negotiation and contracting with suppliers. However, the adoption of e-procurement encounters various exceptions from internal and external environments such as sharply increased demand, delivery delay and inventory failure. In this paper, we have proposed an agent and Web service based architecture for exception handling in e-procurement. Agent technology is applied to deal with the complex, dynamic, and distributed e-procurement process, while Web service technology is applied to provide scalability and interoperability. In this architecture, different tasks in the e-procurement process, such as searching, negotiating, supplier selection, contracting, monitoring, and exception handling, are assigned to different agents, which are wrapped as Web services. A set of rules for detecting and handling the exceptions is defined based on a basic inventory model and a genetic algorithm is utilized for supplier selection. To evaluate our exception handling approach, we have developed a prototype system, through which a simulation has been conducted to verify the effectiveness of our approach.
Electronic procurement (e-procurement), which has been widely adopted to facilitate the acquisition of commodities through the Internet (Barua, Konana, Whinston, & Yin, 2001), mainly focuses on automating the major steps of a purchasing process, such as definition of product requirements, search and selection for suppliers, negotiation and contracting with suppliers (Lancioni, Schau, & Smith, 2003). Companies apply e-procurement technologies for the purpose of reducing administrative costs, cutting down the order fulfillment cycle time, lowering inventory levels and the price paid for goods, and preparing organizations for collaboration with others (Croom, 2000). Through an e-procurement system, the suppliers’ product descriptions and prices can easily be communicated to buyers. Buyers can also submit purchase orders electronically to suppliers through e-procurement. Given the great potential of e-procurement, it is important to study both business and technical issues in implementing e-procurement (Rajkumar, 2001).
Challenges still remain in the application of e-procurement, however, since uncertainties resulting from demand, process, and supply, such as demand fluctuation, delivery delay, and inventory failure, can cause exceptions in the procurement process (Chopra and Meindl, 2003, Ho et al., 2005 and Mason-Jones and Towill, 1998). When such exceptions occur, an e-procurement system should be able to automatically gather relevant information and respond to the exceptions by identifying appropriate alternative suppliers. Thus, the flexibility to handle exceptions is critical for any e-procurement system. Although the existing research has investigated the impact, risks, and uncertainties for adopting e-procurement in an enterprise environment, limited support is offered to handle exceptions in e-procurement management.
Agent technology can automate the execution of complex tasks and has been proposed as a technology for computer-based decision support in the procurement domain (Nissen & Sengupta, 2006). In this paper, we propose a multi-agent based approach to deal with the uncertainties encountered in a procurement process. In our architecture, different tasks in the procurement process, such as searching, negotiating, contracting, supplier selection, exception monitoring and handling, are delegated to a number of both autonomous and collaborative agents. Given that agents can reason using specific knowledge and then react to different situations accordingly, our approach provides a more intelligent, flexible, and autonomous solution to exceptions handling in e-procurement management. Moreover, by wrapping each agent as a Web service, our architecture provides greater scalability and interoperability.
The rest of this paper is organized as follows. Section 2 briefly reviews the relevant literature on e-procurement, intelligent agent, Web service, and agent based applications in e-procurement. In Section 3, we propose an agent based architecture for e-procurement exception management. In Section 4, we develop a prototype and in Section 5 we evaluate our exception handling approach. Finally, Section 6 concludes this paper with our contributions and future research directions.
With the increased complexity, uncertainty, and risks in business transactions, exception handling has become an increasingly important aspect of e-procurement. In this paper, we explore the feasibility to use agent technology for e-procurement exception management. We show that agents that function differently can collaboratively respond to exceptions in a procurement process by detecting the exception, identifying appropriate suppliers, and generating new orders accordingly. We exemplify a set of rules for detecting and handling possible exceptions and a genetic algorithm is developed to select suppliers. In addition, all agents are wrapped as Web services which are reusable for other business applications. Moreover, we have developed a prototype system for e-procurement exception handling. By simulating a restaurant e-procurement process, we have studied the performance of our exception handling approach and the effectiveness of our approach has been verified.
In this paper, the rules for detecting and handling different exceptions are derived from the adopted inventory model. We have applied a basic inventory model, and when a more complex inventory model is used, existing rules need to be modified and new rules may be derived from the inventory model.
