Center Of Excellence Scheduling With Earliness/Tardiness Penalties

Aeroengine has long been recognized as the flower and the most resplendentjewel on the crown of the modern industrial. It is the reperesent of scientifical andtechnological stand of a nation. However, aeroengine is a big burden for thedevelopment of Chinese aviation industry. Process planning and scheduling of theaeroengine manufacturing are the most difficult problem it has been meet, except forthe needed techniques and materials. Recent years, with the increasing demand for theaeroengines, the conflict between the ability and the output of an aeroengine industryhas grown rapidly. In order to find a solution for this conflict, a new kind ofproduction running model for shop floor, center of excellence (COE), has beenpresented and practiced in areoengine industry widely in China.COE is an independent, self and efficient running facilities. It has a high demandfor the due date, lean production and low cost running of the aeroengine production.The dissertation focuses on the key problem of planning and scheduling of COE. Ithopes the research will provide a theoretical foundation and guidance for the practiceof COE in China’s aeroengine industry.The main research works and contributions of the dissertation are as follows:1. For the scheduling of production of aeroengine components in COE withearliness/tardiness penalties, the author sets up a scheduling model which minimizesthe total earliness and tardiness costs. The BOM structure, sequence of parts beingprocessed, and the capability of machines are considened as constraints. The0-1programming method is used to separate the component into parts, which will solvethe problem of quick identification and decision of parents and children in BOM. Anextened genetic algorithm is put forward to solve the problem. Advanced planningand scheduling (APS) is used in simulation and the results show that the method isreasonable and effective.2. Since there is no upper bound for the tardiness of a part in E/T scheduling, theauthor sets up a model for the parts scheduling in COE with deadline constraint. Thestrategy of minimizing tardiness of a part with priority is used. The problem is dividedinto three sub-problems, which are tardiness, repair and earliness problems. So thecomplexity of the problem will be reduced. An extened genetic algorithm (EGA) isused to solve the problem. The active decoding, repair of chromosomes and revesescheduling are used in the decoding of the chromosomes. So the deadline constraintwill be meet as well as the total earliness and tardiness cost will be minimized at the same time.180benchmark simulation cases are studied by taking the size of thescheduling, slackness of due date and the deadline as parameters to be changed.Simulation results show that the model of scheduling and the algorithm have goodability in optimization.3. Some parts of aeroengine are very complicated in manufacturing since theyhave many operations to be processed and they are time consuming. A schedulingmodel of minimize total earliness and tardiness cost with every operation in itsprocessing is set up. A hybrid algorithm of Tabu search and mathematicalprogramming is presented to solve the problem. During the Tabu search, theoptimized operations’ sequences are obtained based on the searching in the swap andinsertion neighboards. So the hardest constraints of machines are staistified. Themathematical programming is used to decide the start time, idle time of the machinessubsequently.72simulation cases are used for the benchmark tests. Simulation resultsshow that the hybrid algorithm of Tabu search and mathematical programming iseffective in solving the problem.4. Taking one of the COE of an aeroengine enterprise in China as an example,the author has put the methodology presented above into practice with the help ofWorkshop Manager, which is a platform of planning and scheduling of COEdeveloped by ourselves. Running results show that the methodology is effective inpractice.