روش شناسی برای ارزیابی پروسه و برنامه ریزی تولید

This paper presents a new methodology for the evaluation of process and production planning activities for decision-making within a computer integrated planning environment. The planning tasks of large production systems involve the management of large amounts of dynamic data, which has to be analysed for decision-making. Through the proposed methodology we aim to improve the effectiveness of both the performed evaluation and the decision-making tasks undertaken when dealing with the different ranges of plans required to satisfy shop floor dynamic requirements and objectives.

Manufacturing planning decisions are made at different phases of the product and process development cycle, such as process planning, master production scheduling, capacity planning and production order scheduling. Actual decisions about the production process are delivered to the shop floor in the form of process plans and production schedules. The integration of planning functions has been cited as a key issue in state-of-the-art manufacturing systems. Process and production planning in particular, which are two multi-task activities of particular importance in a batch manufacturing environment, are prime candidates for the integration effort. Synergies can be generated by changing the traditional and non-integrated use of these systems into collaborative tasks, where enhanced advantages are obtained from the functions performed by each system in conjunction with the shared use of information. Process and production planning integration will result in shorter lead-times and more efficient manufacturing systems with a high competitive advantage. Traditionally, process planning and production planning are subject to different constraints due to the differing main focus of each of these tasks. Process planning is more concerned with the technological requirements of each job, while production planning and control (PPC) systems are responsible for planning the utilisation of production resources (e.g. machine capacities, labour, production quantities), which are required to satisfy certain performance criteria, over a given planning horizon, taking into account a specific demand pattern. In reality these differences usually result in conflicting objectives and the evaluation of planning tasks is carried out using a single criterion for both process and production planning. The decision-making module has an important role within all integrated system architecture [1]. An effective system for evaluation of alternative plans becomes critical within the overall integrated concept of process and production planning. Thus, the relevance of suitable techniques which allow decision-making problem dimensions to be reduced. This paper presents a new methodology that is being developed for the evaluation and selection of alternative plans, generated within an integrated environment. Despite the nature of the hierarchical structure identified within a factory and the boundaries that can exist in decision problems, we consider the methodology described to be of a sufficiently general character to be applicable regardless of any particular level within a factory.

Lack of confidence in the expected results when a specific strategy is implemented has long been a restriction for an effective management system. Within the context of integrated use of process and production planning, the right evaluation and consequent choice of the best of existing alternative plans can significantly reduce the uncertainty factor. The use of discrete simulation facilities as part of the proposed evaluation module has proven to be very efficient as a data generator and support for the development of statistical analysis. It may also be seen also as a way to integrate the data from each planning task, thus improving decision-making in the planning process. The proposed evaluation methodology uses principal components analysis techniques and allows the creation of a new set of criteria which guarantees that relevant information is not ignored, and that the correlation between the original variables is taken into account. The data represented through each principal component, in distinct clusters, allows choice according to decision-maker preferences and avoids the ambiguity that can exist in the choice of the “best” solution/alternative. This methodology using principal components Analysis has shown that the use of new criteria is worthy of consideration as a way to integrate planning tasks.