發(fā)布時間：2018-08-16 13:19

【摘要】：多項目管理是一種綜合性管理活動,它涉及項目的并行運作與項目之間對資源的爭奪,以整體利益最大化為目標(biāo)。在船舶建造過程中,分段制造涉及復(fù)雜的工序,繁多的資源,單靠人力進行編制計劃已經(jīng)不太現(xiàn)實。多項目執(zhí)行下產(chǎn)生的沖突,對船舶企業(yè)的項目管理能力提出了更高的要求。通過多項目視角,對船體分段制造過程進行研究,有利于從整體把握各個項目間進度。通過有效的調(diào)度算法,對提高企業(yè)效率有著重大意義。首先對船舶分段制造流程進行了詳細的分析,特別考慮在分段制造過程中的最重要資源場地資源。在此基礎(chǔ)上總結(jié)了其中存在的約束條件,對工序約束,場地約束,信息傳播約束進行了分析。同時從另一方面介紹了MAS(Multiple-Agent System)的相關(guān)理論,它被應(yīng)用在解決復(fù)雜問題中。本文結(jié)合多項目的問題特點,引入MAS建模思想,建立起基于MAS的多項目場地資源調(diào)度模型,其中針對全局目標(biāo)與局部目標(biāo)之間的矛盾進行了討論。對項目內(nèi)的角色進行了封裝,對各Agent進行了定義。其次,在提出問題模型之后。針對多項目問題解決分三個階段:第一個階段是單項目局部調(diào)度:通過運用貪婪自適應(yīng)算法與可變鄰域技術(shù)的結(jié)合,尋找對單項目較優(yōu)的調(diào)度方案。第二個階段是調(diào)整:利用MAS中所定義的中介收集匯總信息,從整體對各個項目計劃進行調(diào)整,生成多項目調(diào)度計劃。第三個階段是協(xié)調(diào):針對多項目下多種調(diào)度方案,也稱作合約。運用協(xié)調(diào)機制中的單文本協(xié)商機制進行協(xié)調(diào)優(yōu)化。它為Agent之間的協(xié)調(diào)與合作提供了較為合理的溝通協(xié)商方式,使整個系統(tǒng)具有一定的全局優(yōu)化能力。最后運用實例說明在多項目下,能夠滿足大多數(shù)項目要求,從整體達到一個平衡的計劃。
[Abstract]:Multi-project management is a comprehensive management activity, which involves the concurrent operation of the project and the competition for resources between the projects, with the goal of maximizing the overall benefits. In the process of ship building, piecewise manufacturing involves complicated processes and many resources. It is not realistic to draw up the plan by manpower alone. The conflicts arising from the execution of multi-projects put forward higher requirements for the project management ability of the shipbuilding enterprises. From the view of multiple projects, this paper studies the process of hull segmenting manufacturing, which is helpful to grasp the progress of each project as a whole. It is of great significance to improve the efficiency of enterprises through effective scheduling algorithm. Firstly, the ship segment manufacturing process is analyzed in detail, especially considering the most important resource, site resources, in the process of subsection manufacturing. On this basis, the existing constraints are summarized, and the process constraints, site constraints and information transmission constraints are analyzed. At the same time, the related theory of MAS (Multiple-Agent System) is introduced, which is applied to solve complex problems. In this paper, a multi-project site resource scheduling model based on MAS is established by introducing the idea of MAS modeling according to the characteristics of multi-project problems. The contradiction between global and local objectives is discussed. The roles in the project are encapsulated and each Agent is defined. Secondly, after putting forward the problem model. There are three stages to solve the multi-project problem: the first stage is local scheduling of single project. By combining greedy adaptive algorithm with variable neighborhood technology, the optimal scheduling scheme for a single project is found. The second stage is adjustment: the mediation defined in MAS is used to collect the summary information and adjust each project plan from the whole to generate the multi-project scheduling plan. The third stage is coordination: multiple scheduling schemes for multiple projects, also known as contracts. The single text negotiation mechanism in the coordination mechanism is used to optimize the coordination. It provides a reasonable way of communication and negotiation for the coordination and cooperation between Agent and makes the whole system have a certain ability of global optimization. Finally, an example is given to show that under multi-projects, most of the projects can be satisfied, and a balanced plan can be achieved from the whole.
【學(xué)位授予單位】：哈爾濱工業(yè)大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2015
【分類號】：U673;F426.474

【參考文獻】

相關(guān)期刊論文 前5條

1 何育靜;劉樹青;;我國造船企業(yè)競爭力研究[J];船舶工程;2015年01期

2 徐賜軍;李愛平;劉雪梅;;基于資源推拉技術(shù)的多項目調(diào)度算法[J];計算機集成制造系統(tǒng);2010年06期

3 劉傳茂;船體分道建造技術(shù)應(yīng)用與發(fā)展[J];造船技術(shù);2003年01期

4 壽涌毅;傅奧;;多目標(biāo)資源受限項目調(diào)度的多種群蟻群算法[J];浙江大學(xué)學(xué)報(工學(xué)版);2010年01期

5 黃惠娟;鹿翠;許慶;;建筑施工企業(yè)多項目資源優(yōu)化的模糊層次分析[J];項目管理技術(shù);2014年03期

，

本文編號：2186103