Research On The Monitoring And Control Methods For The Projects Of Mould Development

Mould manufacture is a typical one-of-a-kind procedure. Along with higher competition in the present world market and greater diversity of customer requirement, supervising the projects become more and more complicated during the period of mould production. The characteristics in the mould manufacturing, include dynamic bill of material, variety of requirement on the resource kinds, the difficulty of estimating the project makespan accurately, the frequency of engineering changes and etc., and then increase the difficulty of progress control of the projects, thus enlarging the urgency of enterprise requirement of a management system for the project monitoring and control.Monitoring and rescheduling projects belong to a typical kind of RCPSP (Resource Constrained Project rescheduling Problem). Under the constraints of the activity-relation and resources, with the help of this model, manager can reschedule the start-time and finish-time of each activity in all the projects sharing common resources to assure their delivery times. However, the research papers are quite few known about the project scheduling problem under uncertainty, and most of them are still focused on the scheduling problem with certain critical path in relatively simple decision-making paradigms. The theory of Markov Decision Process (MDP) is one of the typical approaches to solve the multi-stage decision-making problems of the project under uncertainty. Therefore, in accordance with the characteristics of the mould project, MDP is employed to create an evolutionary mould which can monitor the project progress with probably multiple critical paths. Based on this monitoring model, another model is also built up to aid the decision-making on the common sharing resources among the multi-projects, thus minimizing the total manufacturing cost.This paper is focused on the monitoring and evolutionary model for the project implementation and the decision-making on the common sharing resources among multiple projects. The research achievement is described as follows: an evolutionary and monitoring model for the project progress is first put forward by means of MDP, providing the probability distribution and the expectation of the project makespan. Secondly, a decision-making model is put forward for the.multi-projects based upon such monitoring model considering the characteristics of the mould projects, thus bringing about total cost as the performace index to optimize and setting up the control paradigm. After that, a stagewise heuristic strategy is employed to obtain a suboptimal decision-making and reschedule for these projects. On the basis of such a model, a module of the computer system is developed to reschedule the projects for the enterprise, At last, a practical case study is performed with this module and a suboptimal reschedule is then obtained to help the managers in the enterprise, showing the effectiveness on monitoring and decision-making in the rescheduling the progresses of the projects.