The Research And Design Of Mould-repair Process Plan System

In the mould-making industry, the competition is largely determined by the technologies and management, experienced engineers and managers are the scarce resource for every enterprise. The production paradigm of the mould is typically based upon the multi-project management with common sharing resources. Experience and knowledge are fully used through all the entire process of mould manufacturing. However, when these persons are retired or pursue higher positions in other companies, the knowledge that they have owned may be lost for the ex-employer. How to preserve the knowledge is an important problem for all the mould manufacturers in the industry. Mould-repair is an important task in the mould production and depends especially upon the experience of the engineer. Now, how to deposit the knowledge on the mould-repair operation will improve its quality, thus reducing the cost and shortening the make span.Knowledge on the mould-repair is much full of subjectivity and uncertainty. For these problems and characteristics above, we start with solidifying mould-repair knowledge and improving its use, and make research and analysis on mould-repair flow and its experienced knowledge, and make a thorough discussion on the compositions and their relationships. Then we propose a method on how to solidify mould-repair knowledge and improve its application. We use Bayesian Networks to represent the uncertainty of mould-repair knowledge and construct a model, which can help us analyze the causes of the fault on the injection and recommend the effective operation for the mould-repair. The main contents include the following work:Firstly, we have performed deep investigation in the enterprise, and analyzed the flow of mould-repair, and discovered some problems of mould-repair and its characteristic. Secondly, we tidy up a lot of case of mould-repair, and then analyze and sum up the compositions for the mould-repair and relationships between them, which lays the foundation for the next step of solving the problem. Thirdly, according to the characteristic of mould-repair, we propose a method by means of Bayesian Networks to solve the problem of depositing mould-repair knowledge and then improve its application. Then we set up a Bayesian Networks model for mould-repair (MRBN), which takes theory and its real application environment into consideration. The practice proves that it can help the engineers analyze the cause of problem well and recommend some more useful solutions. After that, we have researched on inferring arithmetic and inferring process, and proposed a kind of inferring tactic, and then we design and develop the inference engine. At last, base upon the practical analysis and some theoretical research, we design and develop a software system for the mould-repair process planning system (MRPP). This system includes three main modules: editor of the mould-repair knowledge, manager of mould-repair process, and statistic of the mould-repair.To sum it up, according to the method we propose in this paper, the system MRPP has been use in a mould manufacturer successfully, and it is satisfied with the effect.