Modeling of and reasoning about construction specifications to support automated defect detection

As several different studies have pointed out, the construction industry faces the problem of errors occurring frequently on construction sites. Deviations and defects on construction sites need to be identified early on to improve the quality and to minimize future rework costs. The main reasons for construction errors, e. g. human error and changing environmental conditions, are uncontrollable. Therefore, improving the inspection process and the assessment of as-built conditions is critical to prevent defects and their propagation and reduce costs and time for rework. Current inspection approaches are error-prone since they rely on inspectors' experiences. As studies have shown, different inspectors inspecting the same construction site can assess the project differently, sometimes overlooking critical issues. Main reasons for this are the different experiences of inspectors and the large amount of project information, including project related construction specifications, which they need to consider during inspection. There is a need for a formalized and automated inspection approach that will enable inspecting construction sites more thoroughly and reasoning about construction specifications more systematically. This includes helping to collect information, to manage the collection process, to reason about the collected information with respect to the construction specifications, and to identify deviations and construction defects. With new reality capture technologies available, like laser-scanners and embedded sensors, new possibilities for collecting large amounts of information from construction sites are evolving. Laser scanners can be used to collect 3D information of the construction environment and embedded sensors can help in collecting and monitoring material information. However, the evaluation of the data collected using these devices is still a critical task and is currently being done manually. This has three major limitations: (1) Inspectors still focus on only a small sample set of the as-built data available at a given time, which can result in overlooking some errors; (2) It still relies on inspectors' knowledge about specifications and on their previous experiences; (3) The manual evaluation of the data is time consuming. To get the most out of the collected information, there is a need for an automated defect detection system that is capable of reasoning about specifications, design information and the collected as-built information. This automated defect-detection system should be able to model and reason about all relevant project related design, schedule, construction specification and as-built information to identify possible deviations in the as-built environment. The system, finally, should be able to provide inspectors with information on construction defects identified automatically. In this research a construction specification representation and reasoning framework was developed to enable automated evaluation of applicability of construction specifications while leveraging information available in semantically-rich project models. Furthermore, the framework allows for extraction of requirements imposed by construction specifications so that they can be made available for further processing to other mechanisms, such as inspection planning and defect detection mechanisms. The approach developed is validated for cast-in-place concrete-related specifications through computational experiments in an artificial test bed and deployment in two retrospective case studies conducted on real construction projects. During the validation, requirements to project modeling specifications, project models, and project modeling tools were identified, that have to be met to support automated reasoning about construction specifications.