Hybrid Petri Nets and Metaheuristic Approach to Farm Work Scheduling

Scheduling problems for general cases are characterized as NP hard, and the computation time required to obtain the optimal schedule will grow exponentially with the problem size. The scheduling problems that consider the limited or shared resources, alterable constraints or environmental changes become very complex in both formulation and solution. Since the solution for these problems has great serviceability and reliability against environmental changes, much research has been devoted in optimization strategy in the presence of a wide range of uncertainties (Li & Ierapetritou, 2008). Such research with application is applicable to not only the manufacturing in industry, but also production in agriculture. Modeling and scheduling in the agricultural domain may be more promising because of the requirement of new approaches to handling the uncertainties in the nature environment. In agriculture, a system that aims to produce maximum amount of profit from available land by high inputs of capital, labour, or efficient usage of machinery, is defined as intensive farming (or intensive agriculture). Like common businesses, many intensive farming units are operating their businesses by the ways to improving profits in farming while reducing costs. In Japan, there are over 190,000 intensive farming units such as farmers’ cooperatives/ agricultural corporations that aim at efficient and large-scale farm management (The Ministry of Agriculture, Forestry and Fisheries of Japan, 2006). These corporations lease and consolidate agricultural lands in vicinal regions, manage large-scale farmland with full mechanization, and carry out farm works entrusted by vicinal farmers. The farmlands managed by these corporations sometimes number over thousands and are scattered within a wide area. In order to gain substantial economic increase and further development, these corporations need to improve the daily work management, extend the contracts of leasing farmland, lease more farmlands, and carry out more extra works. As a consequence, they considerably require wise management decisions such as timeliness in all operations, equipment adjustments, crop rotations, land rent, taxes and so on. The best decision certainly conduces to the increase of yield, profitability, and work efficiency. Solving the farm work scheduling problem requires appropriate approaches to modeling and optimization. There are plenty of mathematical models and approaches have addressed