【摘要】：公交車作為城市公共交通運輸中重要的一員,擔(dān)負著極其重要的角色,其與軌道交通在城市公共交通運輸中是互相補充,互相依存的關(guān)系。公交車要為乘客提供方便、舒適、快捷的出行服務(wù),其中一個重要前提是必須確保運輸車輛有良好的車質(zhì)車況。這樣的要求,依賴于公交維修企業(yè)對運輸車輛有十分到位的維修保養(yǎng)服務(wù)。另一方面,公交車輛的維修保養(yǎng)在公交企業(yè)經(jīng)營管理中成本高達25%,所以,在確保良好車質(zhì)車況的同時,又不得不考慮維修成本等的問題。車輛在日常的維修保養(yǎng)中,會產(chǎn)生大量的維修數(shù)據(jù),依靠數(shù)據(jù)挖掘技術(shù),運用一些方法和算法,找到隱藏其中的知識,用于優(yōu)化公交車輛日常維修保養(yǎng),為每一輛車都制定差異化的維修保養(yǎng)方案,使公交車輛在維修質(zhì)量、維修效率和維修成本中找到一個合理的平衡點,以提升公交車輛服務(wù)乘客的質(zhì)量,提高公交企業(yè)的經(jīng)濟效益。為解決上述提出的問題,本文首先通過分析公交車輛維修保養(yǎng)信息管理系統(tǒng)生成數(shù)據(jù)的構(gòu)成及特點,結(jié)合公交車輛維修企業(yè)管理的實際需要,提出了公交車輛維修保養(yǎng)數(shù)據(jù)挖掘模型。然后介紹了序列模式挖掘相關(guān)理論、方法和算法,簡要分析了各種算法的優(yōu)劣、應(yīng)用環(huán)境。經(jīng)過對比與分析,本文選定了Apriori算法和FP-Growth算法作為上述模型的數(shù)據(jù)挖掘算法,并在第4章中對這兩種算法進行了較為詳盡的說明和實現(xiàn)。接著,在第5章中提出了本數(shù)據(jù)挖掘模型的技術(shù)線路圖,并對技術(shù)線路圖中問題定義、數(shù)據(jù)準(zhǔn)備、數(shù)據(jù)選擇、數(shù)據(jù)預(yù)處理、數(shù)據(jù)轉(zhuǎn)換、數(shù)據(jù)挖掘、可視化模式規(guī)則和知識庫生成等各個步驟進行了詳細的說明,完整地介紹了公交車輛維修保養(yǎng)數(shù)據(jù)挖掘分析模型的實現(xiàn)過程。然后,根據(jù)數(shù)據(jù)挖掘模型輸出的可視化模式規(guī)則生成了對應(yīng)的知識數(shù)據(jù)庫,該數(shù)據(jù)庫能動態(tài)應(yīng)用于公交車維修保養(yǎng)信息管理系統(tǒng)中。論文還對算法運行的結(jié)果進行了較為詳盡的比較分析,說明了FP-Growth算法對比Apriori算法無論是從算法的執(zhí)行效率,還是系統(tǒng)開銷進行比較,都優(yōu)于不少。最后,論文對公交車輛維修保養(yǎng)數(shù)據(jù)挖掘模型適用性進行了描述,即說明了該模型不但對車輛的報修數(shù)據(jù)能挖掘出可靠的知識數(shù)據(jù)庫,還能對車輛維修數(shù)據(jù)和零配件使用數(shù)據(jù)進行挖掘。論文的結(jié)論部分,從模型的層級設(shè)計、挖掘算法、技術(shù)線路圖和知識數(shù)據(jù)庫使用等方面進行了評價,并得出了該數(shù)據(jù)挖掘模型能基本滿足公交維修企業(yè)使用要求的結(jié)論。
[Abstract]:As an important member of urban public transportation, bus plays an extremely important role, and it and rail transit complement and depend on each other in urban public transportation. In order to provide convenient, comfortable and fast travel service for the passengers, one of the important prerequisites of the bus is to ensure that the transport vehicle is in good condition. Such requirements, rely on public transport maintenance enterprises to transport vehicles in place to repair and maintenance services. On the other hand, the maintenance of public transport vehicles in the operation and management of public transport enterprises costs as high as 25%, so in order to ensure good vehicle quality, but also have to consider the cost of maintenance and other issues. In the daily maintenance of vehicles, a large number of maintenance data will be generated, relying on data mining technology, using some methods and algorithms to find the hidden knowledge, which can be used to optimize the daily maintenance of public transport vehicles. In order to improve the service quality of public transport vehicles, a differentiated maintenance plan is made for each vehicle to find a reasonable balance among the maintenance quality, maintenance efficiency and maintenance cost. In order to improve the quality of bus service to passengers, a reasonable balance can be found among the maintenance quality, the maintenance efficiency and the maintenance cost of the public transport vehicles. Improve the economic benefits of public transport enterprises. In order to solve the above problems, this paper first analyzes the structure and characteristics of the data generated by the public transport vehicle maintenance information management system, combined with the actual needs of the public transport vehicle maintenance enterprise management. A data mining model for bus vehicle maintenance is proposed. Then, the related theories, methods and algorithms of sequential pattern mining are introduced, and the advantages and disadvantages of these algorithms are briefly analyzed. After comparison and analysis, this paper selects Apriori algorithm and FP-Growth algorithm as the data mining algorithm of the above model, and in chapter 4, the two algorithms are explained and implemented in detail. Then, in chapter 5, the technical circuit diagram of the data mining model is proposed, and the definition of technical circuit diagram, data preparation, data selection, data preprocessing, data conversion, data mining, Various steps such as visual pattern rules and knowledge base generation are explained in detail. The implementation process of the data mining and analysis model for bus vehicle maintenance data is introduced in detail. Then, the corresponding knowledge database is generated according to the visual pattern rule output from the data mining model, which can be dynamically applied to the bus maintenance information management system. This paper also makes a detailed comparison and analysis of the results of the operation of the algorithm, which shows that the FP-Growth algorithm is better than the Apriori algorithm in terms of the efficiency of the algorithm and the cost of the system, both in terms of the efficiency of the algorithm and the cost of the system. Finally, the paper describes the applicability of the data mining model for bus maintenance, that is, the model can not only mine a reliable knowledge database for vehicle repair data. Also can carry on the mining to the vehicle maintenance data and the spare parts use data. In the last part of the paper, the hierarchical design of the model, the mining algorithm, the technical circuit diagram and the use of the knowledge database are evaluated, and the conclusion that the data mining model can basically meet the requirements of the public transport maintenance enterprises is obtained.
【學(xué)位授予單位】：華南農(nóng)業(yè)大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2016
【分類號】：U472;TP311.13

【參考文獻】

相關(guān)期刊論文 前1條

1 張國鳳;劉望球;;某型公交車二級保養(yǎng)決策優(yōu)化[J];公路與汽運;2010年04期

，

本文編號：2272054