Scalable Enterprise Systems: Procurement Problem Solving Modeling in Futuristic Enterprise Supply Chains Using Multi-Agents, Stochastic Programming and Game Theory

制造公司需要就选择提供材料的供应商和生产最终产品的制造地点做出决定。供应链中的这种决策问题被称为采购问题。 采购问题的解决方案主要包括供应商和制造地点之间以及制造地点和客户之间的最低成本运输。 本研究针对制造业使用货运公司作为第三方运输工具时的采购问题。 解决的基本问题是：（1）最小化的投标（运输成本）的制造公司，（2）最大化的回报，货运公司，（3）成本最优和成本可靠的估计的任务，由货运公司的车辆分配，和（4）可靠的履行订单的车辆司机。 本研究以多智能体范例、博弈论与随机规划为工具，探讨问题求解模式、沟通机制与求解过程。 该方法将使用司机的本地行为沿着旅行时间和服务时间获得实时通过车载传感器确定的瓶颈上的可行路线，并宣布合作请求contingences.This工作将提供一个可扩展的信息密集型建模和实时采购问题解决的解决架构的基础。 这项研究的结果可以扩展到建立一个集成的供应链基础设施（实时），在电子商务的世界。

Manufacturing companies need to make decisions regarding the choice of suppliers providing materials and the manufacturing sites for producing the final products. This decision-making problem in a supply chain is called the procurement problem. The solution to the procurement problem mainly consists of the least cost transportation between the suppliers and the manufacturing sites, and between the manufacturing sites and the customers. This research project tackles the procurement problem when the manufacturing company uses freight companies as third party transportation. The fundamental issues addressed are: (1) minimization of bid (transportation cost) for the manufacturing company, (2) maximization of payoff to the freight companies, (3) cost-optimal and cost-reliable estimation of the task-to-vehicle assignments by the freight companies, and (4) reliable fulfillment of the orders by the drivers of the vehicles. The research seeks to develop the problem solving model, communication mechanism and the solution process by using multi-agent paradigm, game theory and stochastic programming. The approach will be to use the driver's local behavior along with travel time and service time acquired in real-time through on board sensors for determining the bottlenecks on a feasible route, and announce collaboration requests for contingencies.This work will provide the basis for a scalable information intensive modeling and problem solving architecture for real-time procurement problem solving. Results of this research can be extended to build an integrated supply chain infrastructure (in real-time), in the e-business world.