PhD Studentship in BIM and 3D Laser Scanning

Applications are invited for a full-time PhD research studentship at Glasgow Caledonian University within the School of Engineering and Built Environment. The studentship is for a period of three years, subject to satisfactory progress and provides payment of tuition fees at the UK/EU rate plus an annual stipend of £14,800 [please note that students from outside the EU are required to pay the difference between International and EU fees, currently this would amount to £7,100 per annum]. The successful candidate will carry out up to 6 hours of academic-related work per week as part of their research training.

Research Discipline
Built Environment

Description

Background: It is well established that the as-designed model (typically BIM these days) is not a true reflection of the built asset at the handover stage simply because of change requests, lack of precise implementation of the design intent and several other issues which creep into the construction of a building. Therefore, when the building is about to be handed over, there inevitably are discrepancies between the as-designed and as-built information models. There is no effective method of capturing the as-built information and integrating it with the as-designed information model at the moment. However, more recently, 3D laser scanning technology has been gaining ground for capturing as-built information. Although this is an encouraging development, there are several challenges that need to be overcome before the laser scanned data can be mapped seamlessly on to the as-designed building information model. These challenges arise due to several reasons. Some of the key reasons are (1) gaps in the point cloud due to occlusions etc. in the building geometry, (2) mapping each element in the point cloud with the corresponding element in the as-designed model, (3) working out a way of recognising the discrepancies between the corresponding elements in the point cloud and the as-designed model, (4) a way of adjusting the as-designed model to eliminate the discrepancies etc. These key issues are in addition to the challenge of data cleaning of the massive amount of data contained in the point cloud.

Relevant Research: Development of accurate as-built information model is an active area of research in different parts of the world. The most active groups are based in the USA and more latterly Cambridge University. The PI has strong links with the Cambridge group as well as the leading US group based at Carnegie-Mellon University. The proposed work is a follow on from the state-of-the-art developments in these groups.

Methodology: The methodology adopted for this work will be a positivist approach and will consist of a series of experiments by way of prototype development to implement and validate the algorithm(s) developed for achieving the objectives of the project. The main tasks to be performed will be (1) to identify features to be used in mapping building components embedded in the point cloud with the corresponding ones in the as-designed BIM, (2) identifying the discrepancies between the corresponding components, (3) making the appropriate changes in the as-designed components to eliminate the discrepancies, (4) develop an algorithm to automatically/semi-automatically do the three tasks (1), (2) and (3); (5) implement a prototype to validate the algorithm. A lot of material already exists for tasks (1), (2) and (3) but this work will carry out an evaluation of the existing body of knowledge in these areas to develop its own framework for developing accurate as-built information models.

Nr of positions available : 1