本文選題：企業(yè)資源計劃　切入點：深度強化學(xué)習(xí)　出處：《自動化學(xué)報》2017年09期

【摘要】：傳統(tǒng)的企業(yè)資源計劃(Enterprise resource planning,ERP)采用靜態(tài)化的業(yè)務(wù)流程設(shè)計理念,忽略了人的關(guān)鍵作用,且很少涉及系統(tǒng)性的過程模型,因此難以應(yīng)對現(xiàn)代企業(yè)資源計劃的復(fù)雜性要求.為實現(xiàn)現(xiàn)代企業(yè)資源計劃的新范式,本文在ACP(人工社會(Artificial societies)、計算實驗(Computational experiments)、平行執(zhí)行(Parallel execution))方法框架下,以大數(shù)據(jù)為驅(qū)動,融合深度強化學(xué)習(xí)方法,構(gòu)建基于平行管理的企業(yè)ERP系統(tǒng).首先基于多Agent構(gòu)建ERP整體建�？蚣�,然后針對企業(yè)ERP的整個流程建立序貫博弈模型,最后運用基于深度強化學(xué)習(xí)的神經(jīng)網(wǎng)絡(luò)尋找最優(yōu)策略,解決復(fù)雜企業(yè)ERP所面臨的不確定性、多樣性和復(fù)雜性.
[Abstract]:Traditional Enterprise resource Planning (ERP) uses static business process design concepts, neglects the key role of people, and rarely involves systematic process models. Therefore, it is difficult to cope with the complexity requirement of modern enterprise resource planning. In order to realize the new paradigm of modern enterprise resource planning, this paper is driven by big data under the framework of ACP( artificial society), Computational experiment (Computational experiment), parallel execution of parallel execution. The enterprise ERP system based on parallel management is constructed by combining the deep reinforcement learning method. Firstly, the framework of ERP integrated modeling is constructed based on multiple Agent, and then the sequential game model is established for the whole process of enterprise ERP. Finally, the neural network based on deep reinforcement learning is used to find the optimal strategy to solve the uncertainty, diversity and complexity faced by the complex enterprise ERP.
【作者單位】： 中國科學(xué)院自動化研究所復(fù)雜系統(tǒng)管理與控制國家重點實驗室;青島智能產(chǎn)業(yè)技術(shù)研究院;中國科學(xué)院自動化研究所北京市智能化技術(shù)與系統(tǒng)工程技術(shù)研究中心;
【基金】：國家自然科學(xué)基金(71702182,71472174,71232006,61533019,61233001,71402178) 復(fù)雜系統(tǒng)管理與控制國家重點實驗室優(yōu)秀人才基金(Y6S9011F4E,Y6S9011F4H)資助~~
【分類號】：F272.9;TP18
，

本文編號：1673133