Tracking and abstracting construction site documentation

This paper addresses how best to capture the as-built story of construction projects through site documents so that construction professionals can effectively track what happened during the construction process, solve problems, pursue valid claims, and extract lessons learnt for use on future projects. At present in the construction industry, only a subset of site documents are in electronic format, and software packages used to create site records are usually not compatible with each other; such facts make it very difficult and costly to manage all records directly. This study is based on the premise that creating electronic abstractions of key site documents is an approach not only feasible but also practical to capture their essence. This study examines both the strengths and weaknesses of the traditional hard copy practice of the industry, and how this practice can be enhanced using Information Technology such that current system advantages are retained and new benefits are derived. Building on previous works, the audience of this research work consists of construction firms and public agencies that oversee the construction of buildings and civil infrastructure. The methodology used in conducting this study involved a thorough review of the literature, including past theses on related topics, an examination of the state-of-the- art as reflected in current commercial software, and a field study of on site documentation practices based on two periods of site observation with a project management firm in a high rise building project. The results of the research were implemented in a research system directed at the needs of industry, which features a multiple-view description of construction projects. Generic properties and some custom properties for abstracting site documents have been identified. The ability to search through these abstractions has also been explored. The ultimate goal is to be able to query the system for all information within the multiple view representation pertaining to a user specified issue. This includes the identification of the relevant documents, physical components, schedule performance, and problems encountered.Applied Science, Faculty ofCivil Engineering, Department ofGraduat