生產(chǎn)計(jì)劃調(diào)度管理制度-文庫吧資料
2024-11-03 01:03本頁面
【正文】 to maintenances, preschedules and so this paper, we study the flexible job shop scheduling problem with availability availability constraints are nonfixed in that the pletion time of the maintenance tasks is not fixed and has to be determined during the scheduling then propose a hybrid genetic alogorithm to solve the flexible job shop scheduling problem with nonfixed availability genetic algorithm uses an innovative representation method thrdand applies genetic operations in phenotype space in order to enhance the also define two kinds of neighbourhood for the problem based on the concept of critical local search procedure is then integrated under the framework of the genetic flexible job shop scheduling benchmark problems and fJSPnfa problems are solved in order to test the the effectiveness and efficiency of the suggested methodology.《A Hybrid genetic algorithm for nowait job shop scheduling problems》 作者：Jason ChaoHsien Pan, HanChiang HuangA nowait job shop describes a situation where every job has its own processing sequence with the constraint that no waiting time is allowed between operations within any NWJS problem with the objective of minimizing total pletion time is a NPhard problem and this paper proposes a hybrid genetic algorithm(HGA)to solve this plex genetic operation is defined by cutting out a section of genes from a chromosome and treated as a subproblem is then transformed into an asymeetric traveling salesman problem(ATSP)and solved with a heuristic , this section with new sequence is put back to replace the original section of incorporation of this problemspecific genetic operator is responsible for the hybrid doing so, the course of the search of the proposed genetic algorithm is set to more profitable regions in the solution experiemental results show that this hybrid genetic algorithm can accelerate the convergence and improve solution quality as well.。本文提出一種新的自適應(yīng)遺傳算法用以求解Job Shop調(diào)度問題。因此近年來傾向于利用人工智能的原理和技術(shù)進(jìn)行搜索，尋找復(fù)雜問題的較優(yōu)解，特別是以效仿生物處理模式以獲得智能信息處理功能的遺傳算法研究最為深入。根據(jù)數(shù)學(xué)模型和假設(shè)條件，競(jìng)爭(zhēng)驅(qū)動(dòng)的作業(yè)車間任務(wù)調(diào)度目標(biāo)就是尋求使得每個(gè)制造任務(wù)均能達(dá)到綜合目標(biāo)值最小、利益均衡的調(diào)度結(jié)果。一方面，通過真實(shí)系統(tǒng)運(yùn)行過程中產(chǎn)生的參數(shù)同步地對(duì)仿真系統(tǒng)進(jìn)行調(diào)整，可以大大提高仿真的準(zhǔn)確性、時(shí)效性、智能化；另一方面，通過仿真提供的數(shù)據(jù)同步地為真實(shí)系統(tǒng)地運(yùn)行提供決策支持，這些將大大地?cái)U(kuò)展仿真系統(tǒng)的應(yīng)用能力。不僅下層的調(diào)度方案建立在上層調(diào)度方案的基礎(chǔ)上；同時(shí)，上層的調(diào)度模塊接受下層調(diào)度方案的局部?jī)?yōu)化解作為啟發(fā)信息引導(dǎo)搜索過程，以加快收斂，在較短的時(shí)間內(nèi)得到全局的優(yōu)化解。其中，仿真模型根據(jù)動(dòng)態(tài)注入的生產(chǎn)數(shù)據(jù)完成自適應(yīng)調(diào)整是整個(gè)DDDAS的核心，本文采用分層優(yōu)化的思想生成調(diào)度方案，同時(shí)達(dá)到全局和局部的優(yōu)化目標(biāo)。仿真系統(tǒng)根據(jù)一定的時(shí)鐘節(jié)拍采樣生產(chǎn)線上的數(shù)據(jù)；然后，將仿真數(shù)據(jù)與生產(chǎn)線數(shù)據(jù)作比較，使用因素分析Agent分析原因。使用戶可以實(shí)時(shí)地控制仿真的整個(gè)過程，并利用仿真結(jié)果指導(dǎo)生產(chǎn)和數(shù)據(jù)采集過程。由于柔性制造系統(tǒng)生產(chǎn)線由上百個(gè)生產(chǎn)設(shè)備構(gòu)成，如果對(duì)所有的生產(chǎn)數(shù)據(jù)進(jìn)行采集必然影響數(shù)據(jù)的采集效率，所以采用移動(dòng)Agent既縮短時(shí)間又提高數(shù)據(jù)的精確性，同時(shí)還可以根據(jù)需要對(duì)數(shù)據(jù)作一定的預(yù)處理以縮小傳遞的數(shù)據(jù)量。同時(shí)，將調(diào)度因素反饋給數(shù)據(jù)采集策略Agent，由后者按一定的策略完成下一步的生產(chǎn)線數(shù)據(jù)采集工作。對(duì)多劇情仿真的管理由仿真劇情管理Agent實(shí)現(xiàn)；同時(shí)，仿真參數(shù)和得到的結(jié)果也將作為知識(shí)保存在案例庫中。與傳統(tǒng)仿真不同，動(dòng)態(tài)數(shù)據(jù)驅(qū)動(dòng)仿真是一種與生產(chǎn)線生產(chǎn)過程并行的仿真方法。優(yōu)化策略管理Agent根據(jù)控制模塊設(shè)定的優(yōu)化目標(biāo)或調(diào)度模塊反饋的優(yōu)化目標(biāo)的達(dá)成情況，在中心推理機(jī)的幫助下，按照推理規(guī)則，進(jìn)行模型和算法的調(diào)整，實(shí)現(xiàn)全局優(yōu)化算法和局部?jī)?yōu)化算法之間的動(dòng)態(tài)協(xié)作，當(dāng)滿足一定的條件時(shí)，返回優(yōu)化的調(diào)度方案作為當(dāng)前的最優(yōu)解。同時(shí)，它也在不斷地動(dòng)態(tài)調(diào)整，保持與生產(chǎn)線當(dāng)前的實(shí)際生產(chǎn)情況一致，完成這個(gè)任務(wù)需要中心推理機(jī)的協(xié)同，如基于案例庫的推理等。各目標(biāo)權(quán)值的大小比例是一個(gè)動(dòng)態(tài)調(diào)適的過程。短期優(yōu)化目標(biāo)包括：最大化生產(chǎn)量、最大化WIP移動(dòng)步數(shù)、最小訂單交貨延遲率、降低加工周期、降低加工周期方差、降低WIP水平等。多目標(biāo)管理Agent：負(fù)責(zé)控制當(dāng)前調(diào)度優(yōu)化的方向，對(duì)調(diào)度方案評(píng)價(jià)函數(shù)中多個(gè)不同目標(biāo)的權(quán)重進(jìn)行動(dòng)態(tài)
研究報(bào)告相關(guān)推薦
鄂ICP備17016276號(hào)-1