Expanding the use of contract inspections in construction : An approach to inter-project knowledge diffusion?

Sammanfattning: This licentiate thesis presents a taxonomic approach to classification of defect information in construction projects. A conceptual model, based in a frame of reference consisting of the concepts of Continuous Improvement, Performance Measurement, Knowledge Management, as well as the current Quality Management regime of the industry helps to understand how the use of Contract Inspections can be developed towards facilitating inter-project knowledge diffusion and continuous improvement in construction.Three research questions were formulated:RQ I:How is information from inspection reports currently used?RQ II:How could information from inspection reports be utilised further?RQ III:How should information from inspection reports be structured and codified, to enable storage and future data analysis to facilitate continuous improvement?Qualitative as well as quantitative data has been collected in three different studies. In the first contractors were surveyed through a questionnaire about their use of different suggested experience feedback concepts, such as contract inspections. In the second other project participants than contractors was an interviewed, asking them how they use inspections and how the current usage could be extended. In the third study, a single case study about the inspection activities in one construction project, the purpose was to analyse the method and results of Pre- and Final Inspections. These results were interpreted through classification theory. It is further suggested how inspection data could be enhanced for the purpose of database storing, transforming, and easy access.The thesis concludes that both contractors and the other project roles currently use and view inspections as they are prompted in the General Conditions (AB 04), i.e. as an activity that document what defects there are for the contractor to rectify. Several companies have started to store inspection reports on project-dedicated servers, a routine that is thought to enhance the access to information to some extent. Several respondents have also introduced the compilation of defect statistics.Except automating the compiling of descriptive defect statistics, a system for managing inspection information, giving the widest access to the information about previously experienced defects, is expected to serve a Continuous Improvement process with input for identifying reoccurring problems in the production process. This system could as well be utilised for giving performance feedback to contractors and suppliers. Through continual surveying of the performance of current projects, the performance of the production process can be monitored. Based on the analysed reports, and the literature on taxonomy, the conclusions for RQ III identifies 15 different types of data tags, or types of information, that should be used as structure for the defects’ information: Unique Project Identifier, Unique Defect Identifier, Inspection Identifier, Responsibility, Defect Serial Number, Floor level, BSAB 96 Space Code, BSAB Object code, Defect type, Defect description, Rectification measure, Flag for Safety Issue, Cause, Date logging, and Photos.Future research should focus on the validation of the proposed model and system, suggested through case studies.