近年隨資訊科技的進步，鋪面檢測採自動化方式收集資料，增進鋪面資料的收集效率，並將大量的鋪面資料儲存於鋪面管理系統的資料庫中，造成資料庫快速的成長，越來越多的資料將為各公路主管機關或研究單位，面對鋪面養護或修復決策時帶來新的挑戰。資料採礦是個整合性的新技術，已擴展應用於資料庫系統、機器學習、智慧型資訊系統、統計與專家系統等學門領域，該技術對於提昇鋪面管理系統決策過程的正確性與效率深具潛力，本研究主要目的係依據資料採礦的觀念，針對中壢工務段轄區道路養護資料，應用分類規則、約略集合理論以及關聯分析法，並探討鋪面養護壽命，以系統化方式將其應用於鋪面管理資料上，萃取有用的鋪面資訊。 分類規則即是研究已經分類好的資料之特徵，再依這些特徵預測其他未分類或新的資料。本研究採C5.0決策樹演算法，針對中壢工務段鋪面管理資料庫進行分類分析，找出關鍵影響鋪面養護維修方式的決策規則與因子，並與約略集合理論分析結果比較。本研究亦採用Aproiri關聯演算法分析柔性鋪面破損因子間之關聯性，並以專家問卷方式輔證本研究分析結果，提升分析結果可信度。最後利用工務段於鋪面養護管理作業之紀錄表單，來探討鋪面養護延長壽命狀況。 本研究以實際資料進行研究，證實資料採礦技術對於分析不精確或不確定之鋪面管理資料是個有用且有效率的工具，將資訊科技的概念應用於鋪面管理上，盼能藉此獲得寶貴的鋪面知識，並協助國內公路主管機關將鋪面管理系統資料庫中的資料作最佳化的運用。 The process of collecting pavement data has been evolving with advances in technology, thus generating huge amounts of data by both objective and subjective methods to be stored in pavement management systems (PMS) databases. The rapid size increases of these databases presents a challenge for state agencies, as they attempt to understand and take advantage of the data to support pavement maintenance and rehabilitation (M&R) decisions. Data mining is an interdisciplinary research area spanning several disciplines such as database systems, machine learning, intelligent information systems, statistics, and expert systems. This approach has the potential to further increase the accuracy and efficiency of the decision-making process for PMS. The primary objective of this study is to apply a systematic data minig technique approach on a large practical pavement management database to extract useful information. The study provides evidences showing that data mining technique constitutes a sound basis for data mining applications and also can be a useful tool for the analysis of inexact, uncertain, or vague pavement data. Study was being conducted in which results from the whole set of data is presented and interpreted in order to obtain a better view of the condition of pavements and be able to increase the effectiveness of the decision-making process.
