營建專案人力配置最佳化及參與時序模式建立;THE IMPACT FACTORS AND PREDICTING ALLOCATION OF ON-SITE AND OFF-SITE MANPOWER FOR CONSTRUCTION PROJECTS

NCUIR > Engineering College > Graduate Institute of construction engineering > Electronic Thesis & Dissertation > Item 987654321/66073

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/66073

Title:	營建專案人力配置最佳化及參與時序模式建立;THE IMPACT FACTORS AND PREDICTING ALLOCATION OF ON-SITE AND OFF-SITE MANPOWER FOR CONSTRUCTION PROJECTS
Authors:	陳韋向;Chen,Wei-Hsiang
Contributors:	營建管理研究所
Keywords:	營建專案管理;人力資源配置;案例式推理;時間序列;預測模式;Human resource management;Construction Project;Allocation;CBR;Time Series
Date:	2014-07-29
Issue Date:	2014-10-15 17:28:23 (UTC+8)
Publisher:	國立中央大學
Abstract:	營建專案之人事費用支出向來即為營建公司重點議題之一，於專案備標階段即需準確評估，倘無法確估人事費用預算，將直接影響得標專案獲利空間，本研究鑑於過往營建專案人力配置並無明確方法或原則可依循，故欲建立一套可有效推估營建專案人力配置數量及參與時序之模式。首先本研究透過文獻回顧方式瞭解影響營建專案人力配置之主要原因為何以及如何借助資料探勘及人工智慧等方式建立預測模式，爾後再透過專家訪談及改良式德爾菲法建構欲蒐集營建專案案例之資料型式。透過國內數家大型營建公司協助，蒐集台灣近十六年來共計一百二十九筆營建專案工程案例，經由敘述性統計分析可得出影響營建專案人力配置之關鍵因素共有十六項，同時並篩選完整案例共計九十七筆作為日後可供存取之案例資料庫。其後，本研究借助案例式推理(case based reasoning, CBR)及時間序列(Time Series)分析模式建立營建專案工程人力配置預測模式，此模式之特色為可依據營建專案各項表徵資訊及各種估算成本即可預測投入專案之人數，並可得到各類型專業人員特有之參與時序曲線，此曲線將可作為專案配置各專業人員時序之參考。最後本研究將建立之預測模式採用三次交叉驗證方式(Three-Way Cross Validation)進行驗證以佐證本研究之價值，由其結果得知此模式之正確率高達88.65%，表示其已足有參考之價值。本研究並發現十六項關鍵因素中，皆屬專案規劃前期可得知之資訊，故本模式將可於新專案前期規劃時，即可輸入各項資訊預測所需之人力並可建構專案各階段各種專業職能人員參與時序之策略。;In the highly competitive construction industry of Taiwan, a slightly inaccurate estimation of a project can easily cause loss of the bid, especially in erroneously estimating costs and allocation of on-site and off-site manpower that usually offsets the profit gained from the project and even jeopardizes the management processes of the project. This research developed a predict model based on mathematical regression, case-based reasoning and time series to predict costs and allocation of on-site and off-site manpower for construction projects. It is founded on laborious processes of data collections and analyses by matching statistical assumptions, and includes construction projects of residential buildings, industrial office buildings, commercial buildings and industrial construction. In this research, 97 cases data filtered by 129 Construction projects cases with Descriptive Statistics method would be used as proper data for analyzing via case based reasoning and Time Series, to establish an optimization model. After verifying by Three-way Cross Validation, the accuracy rate of this model achieved 88.65%. This research can be used in project bidding stage for estimating on-site and off-site manpower of each position of whole project life cycle.
Appears in Collections:	[Graduate Institute of construction engineering] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	502	View/Open

社群 sharing

Loading...