【摘要】：煤炭作為我國(guó)能源消耗的主要來(lái)源之一，在經(jīng)濟(jì)發(fā)展當(dāng)中具有舉足輕重的地位。隨著信息技術(shù)的不斷發(fā)展，互聯(lián)網(wǎng)技術(shù)和移動(dòng)互聯(lián)網(wǎng)技術(shù)的廣泛應(yīng)用，以及經(jīng)濟(jì)全球化進(jìn)程的加劇，使得市場(chǎng)競(jìng)爭(zhēng)更加激烈。國(guó)內(nèi)的煤炭企業(yè)物資計(jì)劃管理信息化水平整體還比較低，隨之帶來(lái)的問(wèn)題是庫(kù)存積壓嚴(yán)重，占用大量資金，企業(yè)生產(chǎn)成本增大。為實(shí)現(xiàn)“零庫(kù)存”的管理思想，延伸供應(yīng)鏈到采煤工作面，做好煤炭企業(yè)物資計(jì)劃的精細(xì)化管理，，利用信息技術(shù)科學(xué)有效地對(duì)煤礦生產(chǎn)所消耗的物資進(jìn)行動(dòng)態(tài)預(yù)測(cè)、建立物資計(jì)劃提報(bào)的智能化管理系統(tǒng)是降低企業(yè)庫(kù)存成本，提高物資供應(yīng)鏈的精細(xì)化管理水平的關(guān)鍵手段。 本文首先分析了當(dāng)前煤炭企業(yè)物資計(jì)劃管理存在的一些問(wèn)題，然后闡述了物資計(jì)劃管理的相關(guān)內(nèi)容，根據(jù)已經(jīng)分類(lèi)好的物資，通過(guò)提取煤礦井下生產(chǎn)作業(yè)所需的這類(lèi)物資的影響因素，設(shè)計(jì)關(guān)鍵指標(biāo)進(jìn)行分析，確立了物資需求預(yù)測(cè)關(guān)鍵指標(biāo)體系。同時(shí)針對(duì)支持向量機(jī)模型在參數(shù)選取時(shí)具有一定的主觀性和參數(shù)優(yōu)化程度不夠的問(wèn)題，采用粒子群優(yōu)化算法對(duì)支持向量機(jī)模型的最佳參數(shù)進(jìn)行最優(yōu)選取，然后將最優(yōu)選取的參數(shù)結(jié)果應(yīng)用于支持向量機(jī)對(duì)物資需求預(yù)測(cè)，以實(shí)際的物資采煤機(jī)截齒為例進(jìn)行預(yù)測(cè)，預(yù)測(cè)結(jié)果表明通過(guò)粒子群算法對(duì)參數(shù)優(yōu)化后的支持向量機(jī)預(yù)測(cè)模型提高了預(yù)測(cè)精度。 本論文以實(shí)際參與的企業(yè)課題“山東能源淄礦集團(tuán)供應(yīng)鏈電子商務(wù)系統(tǒng)”為背景，通過(guò)分析淄礦集團(tuán)當(dāng)前計(jì)劃管理的業(yè)務(wù)需求，提出了煤礦企業(yè)物資計(jì)劃管理系統(tǒng)的整體架構(gòu)，探討了系統(tǒng)的技術(shù)實(shí)現(xiàn)方案。對(duì)其中的物資預(yù)測(cè)進(jìn)行數(shù)據(jù)建模，選擇粒子群優(yōu)化的支持向量機(jī)作為物資需求的預(yù)測(cè)模型，基于J2EE開(kāi)發(fā)平臺(tái)，并采用Oracle10g數(shù)據(jù)庫(kù)作為底層數(shù)據(jù)支撐平臺(tái)，同時(shí)以Activiti5流程引擎設(shè)計(jì)并開(kāi)發(fā)了一套符合淄礦集團(tuán)實(shí)際業(yè)務(wù)的基于PSO-SVM(Particle Swarm Optimization-Support Vector Machine，PSO-SVM)的物資計(jì)劃管理平臺(tái)，實(shí)現(xiàn)了淄礦集團(tuán)物資計(jì)劃管理的信息化，最后對(duì)論文的研究工作做以總結(jié)，并對(duì)系統(tǒng)的進(jìn)一步研究做以展望。
[Abstract]:Coal, as one of the main sources of energy consumption in China, plays an important role in economic development. With the continuous development of information technology, the wide application of Internet technology and mobile Internet technology, as well as the aggravation of the process of economic globalization, the market competition is more intense. The information level of material plan management in domestic coal enterprises is still relatively low as a whole, which brings about the problem of serious inventory backlog, occupying a large amount of funds and increasing the production cost of enterprises. In order to realize the management idea of "zero inventory", extend the supply chain to the coal mining face, do a good job in the fine management of the material plan of coal enterprises, make use of information technology to predict the materials consumed in coal mine production scientifically and effectively, and establish the intelligent management system of material plan reporting is the key means to reduce the inventory cost of enterprises and improve the fine management level of material supply chain. This paper first analyzes some problems existing in the material planning management of coal enterprises at present, and then expounds the related contents of the material planning management. According to the classified materials, by extracting the influencing factors of this kind of materials needed in the underground production and operation of coal mine, the key indexes are designed and analyzed, and the key index system of material demand prediction is established. At the same time, in order to solve the problem that the support vector machine model has certain subjectivity and the parameter optimization degree is not enough in the parameter selection, the particle swarm optimization algorithm is used to select the best parameters of the support vector machine model, and then the optimal selection parameter results are applied to the support vector machine to predict the material demand, and the actual material shearer cutting teeth are taken as an example to predict. The prediction results show that the particle swarm optimization algorithm is used to improve the prediction accuracy of the optimized support vector machine prediction model. Based on the actual enterprise project "supply chain Electronic Commerce system of Shandong Energy Zimine Group", this paper analyzes the business requirements of the current planning management of Zimine Group, puts forward the overall structure of the material planning management system of coal mining enterprises, and probes into the technical realization scheme of the system. The data modeling of material prediction is carried out, and the support vector machine of particle swarm optimization is selected as the prediction model of material demand, based on J2EE development platform, and Oracle10g database is used as the underlying data support platform. At the same time, a set of material planning management platform based on PSO-SVM (Particle Swarm Optimization-Support Vector Machine,PSO-SVM is designed and developed with Activiti5 process engine. Finally, the research work of the paper is summarized, and the further research of the system is prospected.
【學(xué)位授予單位】：西安科技大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2014
【分類(lèi)號(hào)】：TP311.52;F426.21;F251

【參考文獻(xiàn)】

相關(guān)期刊論文 前1條

1 李建民,張鈸,林福宗;支持向量機(jī)的訓(xùn)練算法[J];清華大學(xué)學(xué)報(bào)(自然科學(xué)版);2003年01期

本文編號(hào)：2504215