Organizational-Memory-Based Performance Prediction of Procurement and Contractor Selection Methods

[[abstract]]採購合約型式及搭配的選商方法的決定為營建專案業主的重要決策，常受組織因素影響．雖然各種系統都有因本質而存在的優缺點，但實地執行的成果未必與理論推測一致．本研究之目的為發展以組織經驗(即組織記憶)為基礎的採購績效預估模式，作為採購決策輔助，以改進既有作法．範圍則限於有持續興建同類型專案需求的業主機構．在此情況下，專案的目標優先次序較為固定，且以往的合約經驗也比較能夠引用於新的情況．這是有別於文獻中常見的依照專家意見、就非特定類型專案、以對各採購方式的一般性屬性評分為基礎的評估模式，也不同於藉案例分析、以專案、業主、市場等類別變數所建立之採購績效預估模式．它們都忽略個別組織獨特性之影響，也未能提供組織學習與知識管理的機制．本研究擬以有長期性發包需求的某大公部門機構為例，蒐集其近年來使用各種合約所完成的同類型工程案例，建構示範模式．首先探討可能影響績效之因子，包括與環境 /市場、公司內部、專案情況、及合約/選商等有關的變數，以作為模式輸入．而各關鍵績效指標及整體滿意度將為模式輸出．除保留一部份案例資料作為模式測試外，大部份將用於建構模式知識庫．對機構所面臨的專案情況，模式將能預估使用各種合約及選商方案的採購績效．有最佳預估績效的方案即為建議之選項．擬採用類神經網路和案例式推理建構模式，以比較二者的精確度．組織記憶模式對組織學習與知識管理的意義亦將被探討．[[abstract]]Determination of an appropriate procurement method for a construction project along with the method for contractor selection constitutes an important decision of the owner in project planning, which often is influenced by organizational factors. Although each of the procurement systems currently available has its inherent strengths as well as weaknesses, the results of actual implementations may not agree with theoretical estimation. The objective of the proposed research is to develop an organizational experience-based, i.e. organizational memory-based, model for predicting procurement performance that can be used as a decision aid for procurement and contractor selection, in order to improve existing practices. The scope of the study is limited to owner organizations with continuous demand for construction projects involving the same type of work. For such owners, project goals and priorities are relatively fixed and moreover, experiences gained from past contracts are more applicable in new scenarios. The model is different from those commonly seen in the literature that evaluate each procurement method based on the ratings of some general attributes for non-specific type projects according to experts’opinions. Also, it is different from those established from analysis of case studies that estimate a procurement method’s performance with variables relating to categories of project, client, and market etc. They do not take into account the effect of uniqueness of an individual organization, nor do they provide a mechanism for organizational learning and knowledge management. Using a large public-sector organization with a long-term demand for construction projects as an example, the proposed research will collect a number of same-type projects that the firm has completed in recent years through various procurement routes and develop an illustrative model from the data. The first step will be to investigate factors that may have an influence on procurement performance, including variables relating to the environment/ market, firm’s internal conditions, project conditions, and contract/award, which are to be used as model inputs. Key performance indicators as well as degree of overall satisfaction will be model outputs. While a portion of the collected case projects will be set aside for model testing, the bulk of the data will be used for constructing the knowledge base of the prototype model. For given conditions of a project faced by the firm, the model thus built would be able to estimate the performance of the project procured using each possible arrangement of contract and contractor selection method. The procurement alternative with the best performance prediction is the recommended choice. To develop the model, it is proposed to use both artificial neural networks and case-based reasoning systems for a comparison of their estimation accuracy. In addition, the research will explore the implications of the organizational-memory-based model for organizational learning and knowledge management.[[note]]NSC97-2221-E327-03