مدل‌سازي و حل مسيله‌ي زمان‌بندي در يك سيستم توليد سه‌مرحله‌يي مونتاژ با هدف تحويل به موقع محصولات

Modeling and Solving a Three Stage Assembly Production System for On Time Delivery

در اين تحقيق مسيله‌ي زمان‌بندي در يك سيستم توليد سه‌مرحله‌يي مونتاژ بررسي مي‌شود. در اين سيستم انواع مختلف محصولات توليد مي‌شود و هر محصول براي تكميل نيازمند قطعات مختص خود است. ابتدا قطعات در يك ايستگاه با ماشين‌هاي موازي پردازش مي‌شوند و سپس در مرحله‌ي دوم مورد كنترل و عمليات تكميلي قرار مي‌گيرند. قطعات هر محصول پس از تكميل، در ايستگاه مونتاژ (مرحله‌ي سوم) مونتاژ مي‌شود و بدين‌ترتيب محصول مربوطه تكميل مي‌گردد. هدف عبارت است از كمينه كردن مجموع زودكرد و ديركرد تحويل محصولات. مسيله‌ي فوق جز مسايل NP-hard محسوب مي‌شود و به‌همين دليل ابتدا مدل رياضي مسيله، و سپس الگوريتمي ابتكاري براي حل آن در ابعاد بزرگ ارايه شده است. در نهايت، عملكرد الگوريتم پيشنهادي در حل مسايل متنوع ارزيابي شده است. همچنين براي مسايل كوچك، نتايج حاصل از الگوريتم پيشنهادي با جواب بهينهي حاصل از حل مدل رياضي مقايسه شده است.

In this paper, the scheduling problem in a three stage assembly production system is studied. In this production system, there are two stages; to fabricate and provide the parts, and an assembly stage to assemble the parts and complete the product. Suppose that a number of products of different kinds are ordered to be produced, and each product needs a set of several parts to complete. At first, the parts are processed in the first stage with some parallel machines and, then, they are controlled and trimmed in the second stage. After completion of the set of parts for each product, the parts are assembled into the product in an assembly stage. The main assumptions are that the machines are uniform in the first stage. In addition, the assembly operation of a product will not start until all parts of its product are completed. Finally, if product A is going to be assembled before product B, then, at each stage, the processing of any part for product B does not start before the start of processing all parts for product A. The considered objective is to minimize the amount of earliness and tardiness of the products. At first, the parameters and decision variables are defined and, then, the mathematical model of the considered problem is presented. Since the considered problem is NP-hard, a heuristic algorithm is developed to solve the big sized problems. Finally, the performance of the proposed algorithm is evaluated in solving some different problems. Based on the proposed algorithm, the problem is solved in two phases: phase 1: to determine sequencing of the products, and phase 2: to determine the sequencing and scheduling of the parts for each product. In this evaluation, the result of the proposed algorithm that is compared with the optimum solutions obtained from solving their mathematical model is a small sized problem.