The recurrence of similar problems caused by human errors in urbanization process is common throughout the world. However, the knowledge learnt from these problems should become lessons as references for decision-making to avoid the recurrence of these problems, thus the results of urbanization are sustainable. It is considered of imperative importance to incorporate the lessons experienced into the decisionmaking process in a way that can help foresee the potential problems and take proper measures for addressing the problems. There is little existing study on how previous lessons are mined and incorporated in foreseeing the potential problems in future. The lack of this mining mechanism presents a significant barrier for decision makers to learn from the existing lessons thus to have references of how to make better decisions for future urbanization practices. This paper presents a Lessons Mining System (LMS) to assist in mining lessons experienced from previous practices. The system includes five components, namely, Lessonscase Representation , Lessons-case Store, Lessons-case Retrieval, Lessons-case Application, and Lessons-case Update. LMS can facilitate decision makers to understand what potential problems might occur from their current actions by referring to the lessons experienced previously in similar circumstances. This understanding can help decision makers take preventive measures to mitigate the potential problems. In other words, the use of LMS can send alarming messages to decision makers about what possible problematic consequence may occur thus they can modify their actions before too late. The establishment of LMS is based on Casebased Reasoning (CBR) theory and the similarity matching principles. A demonstration of Yangwu Town is presented to show the application of the system, and the result shows that the lessons mined can provide valuable references for the government of Yangwu Town to improve their decisionmaking quality.