دانلود مقاله Development of an Efficient Machine Loading Heudistic for Cellular Manufacturing System
سال انتشار: ۱۳۸۳
محل انتشار: سومین کنفرانس ملی مهندسی صنایع
تعداد صفحات: ۱۳
Nirjhar Roy – M.N. National Institute of Technology, Allahabad, India
Cellular manufacturing aims at identification of families of components (parts) and their associated machine groups in a job shop or a batch processing system. It is often observed that clear transformations of discrete manufacturing systems to cellular manufacturing systems are not feasible due to various practical constraints. Any discrete manufacturing system becomes dynamic as different sets of parts are loaded at different time periods. Further creating virtual cells will not automatically maximize the utilization of any system unless the set-up times are reduced by sequencing the parts most judiciously. A practical difficulty that has been experienced by the researchers in partand- machine-association is that a single set of tools in a dedicated machine-group cannot process all the parts of its associated part family. Tool changes are generally required. Thus in the cellular manufacturing systems tooling families are always formed and must be identified to exploit the available resources. The present work integrates a clustering algorithm and a group-scheduling algorithm (known as machine loading algorithm) to identify a product-mix strategy for any discrete manufacturing system where facility-timings are limited and all the parts may not get processed within the time limits. The clustering model first creates the (virtual) cells of machine groups. The machine loading algorithm thereafter utilizes an algorithm known as Minimax algorithm which determines the sequences, and the types of parts (of batches) that can be selected and optimally processed when allowable times are less, (more) or equal to the required timings of processing of all the loaded parts. If the available time is less the algorithm selects the parts that should be included and sequenced in a time frame. The machine loading algorithm utilizes another heuristic, SWAP in case a near optimal product-mix solution is resulted after using the Minimax algorithm. The validity of the model has been tested suitably to prove its usefulness. The model also includes realistic situations where machine groups could be changed by utilizing alternate process plans.