Kuttantharappel Soman, R (2020) Automated look-ahead schedule generation using linked-data based constraint checking for construction projects. PhD thesis, Imperial College London, UK.
Abstract
Poor planning on the construction site at the ‘look-ahead’ planning stage (where information from diverse sources is integrated as plans are being developed for the next six weeks) often result on cost overruns and schedule delays. This thesis addresses the inefficiencies emerging from manual look-ahead planning by means of an information modelling approach to codify and validate detailed construction process information. The key contribution is a novel Linked-Data based Constraint Checking (LDCC) method to identify construction constraint violations from information distributed over multiple heterogeneous sources. LDCC can be integrated with machine learning methods to generate a constraint-free Look-Ahead Schedule (LAS) automatically. This approach can augment decision-making in look-ahead planning through data-driven constraint identification and automated LAS generation. This research comprises of three main studies. First, digital information use in three construction projects was studied to inform the development of the information modelling approach. This study identified three codification challenges—software usage, information sharing, and missing construction process information. Second, building on the understanding of the codification challenges, the LDCC method was developed to codify and validate detailed construction process information, including complex construction constraints distributed over multiple databases. Third, the LDCC method was successfully integrated with two machine learning methods (Genetic Algorithm and Reinforcement Learning) to automate LAS generation. When tested on a real construction project, the LDCC method identified all the constraint violations in the manually generated LAS. Also, LAS generation methods automatically generated conflict-free LASs significantly faster than manual methods. Both results demonstrate the applicability of the developed methods on real construction projects. In summary, the thesis extends existing knowledge in the construction informatics domain by demonstrating the benefits of using linked-data based methods to address the issue of data silos in construction information, and enabling the application of data-driven-decision support tools to aid look-ahead planning.
| Item Type: | Thesis (Doctoral) |
|---|---|
| Thesis advisor: | Whyte, J and Molina-Solana, M |
| Uncontrolled Keywords: | construction project; construction site; learning; machine learning; information modelling; scheduling; construction planning |
| Index terms: | cost overrun, information modelling, database, schedule delay, construction planning, reinforcement, information sharing, construction site, informatic, genetic algorithm, scheduling, decision support, machine learning, violation, construction process, decision-making, construction project |
| Subjects: | regulatory law, information science, data management, project controls, decision analysis, data exchange, artificial intelligence, financial and cost management, construction planning, building materials, work location, building construction, production management, digital design, algorithms, operations research |
| Topics: | Construction Materials, Cost Management, Site Management, Time Control, Digital Applications, Risk Management, Project Management, Legal Issues |
| Descriptive scope: | 2 PC |
N.B. Descriptive scope is a count of how many of the five facets of empirical research are indicated by the words used in title, abstract and keywords. It is not intended as a judgement on the research; merely a count of the kind of word we would expect to indicate Phenomenon, Concepts, Theoretical framing, Empirical techniques, Analytical techniques. If all five are present, then a code of “5 PCTEA” will indicate this. If you feel the coding for this record is questionable, we welcome discussion around the terms we matched or the way we categorized them. The facet you would expect may not be coded, or a facet may be coded inappropriately. This can also bear on a larger question, of which facets should be treated as defining in construction management research. Please get in touch, and we will look at it. More details here