Research on Two-agent Scheduling Problem with Hot Lots

Different customers have different requirements on product quality, due date and so on, the company have to weigh these requirements against costs, seeking a win-win strategy which can not only gain customer satisfaction, but also make the maximize business efficiency. The study of multi-agent scheduling could exactly provide a good idea for the solution of the problem. Considering the instability of the market, the paper tries to introduce hot lots into the research of two agent scheduling problem, and seek scheduling methods to accelerate the pace of market responsiveness. This study focuses on the scheduling problem of two competing agents on parallel-batching machine with the introduction of a new agent C as hot lots. We established a mathematic model of two competing agents about hot lots. And we also designed two heuristic algorithms (H1 and H2) to solve the model respectively, one is based on NFD rule and another is based on FFD rule. Then we analyzed these two algorithms performance through theoretical analysis and MATLAB-based simulation experiment. Experimental results show that both algorithms can operate effectively, only several examples are not feasible among the applications generated randomly. However, the performance of the algorithm H1 is relatively optimal and stable, and its running time is significantly less than the algorithm H2.