Dynamic Capabilities: Application of Artificial Intelligence and Knowledge Representation in Management of Engineering Mega projects

Mega projects are the major mode of development across the engineering sectors of infrastructure, energy, and mining. Including other fields with similar delivery models such as defence, aerospace, and global events, it is estimated that 8% of annual world gross domestic product (GDP), or US$6 to $9 trillion dollars, is spent on megaprojects. The infrastructure sector is responsible for nearly half of that figure (Flyvbjerg, 2014).

Capital Expenditure (CapEx) estimates for these projects overrun for a variety of reasons and mechanisms (Hollmann, 2016). Several studies investigate these cost overruns and reported the average cost overrun to be around 20% for road projects, 45% for rail projects (Flyvbjerg et al. 2002), and 22% for mining projects (Gypton, 2002). Recently, a comprehensive investigation reported up to 65% of the industrial megaprojects failed to meet their key objectives with 40% cost overruns on average (Merrow, 2011). Improving these estimates will provide better financing, better project lifecycle management (Hickson and Owen, 2015), and above all, help save private and public resources by identifying better investment opportunities, and strategies to shape projects before sanction.

Large engineering companies possess an abundance of unstructured project data, and valuable expert knowledge. While the data is often stored in isolated silos losing relevance in time, the expert retirement phenomenon in the industry causes valuable knowledge disappear. There is enormous potential to systematically utilise these resources to assist with assessment of risks and estimates.

Advances in computer and data sciences have significantly changed our interaction with data. More devices get connected to the internet each day. “Sematic Web” has raised as a solution to increase machine readability and interoperability of data using “Linked Data” format and graph data structures (Berners-Lee et al. 2001). This provides a great environment to capture knowledge using ontology engineering (Pauwels et al. 2017).

Many algorithms have been proposed and tested to improve CapEx estimates (Emsley et al. 2002, Lowe et al. 2006, An et al. 2007, Abdelgawad and Fayek, 2010). However, the complexity of megaproject’s domains requires explicit expert knowledge of causalities to draw meaningful inferences. Probabilistic graphical models (PGM) are a branch of artificial intelligence that allow such casual relations (Koller and Friedman, 2009). The applications of Bayesian Belief Networks (BBN) as a PGM method are proved successful in modeling in engineering projects cost and schedule (Nasir et al. 2003, Khodakarami and Abdi, 2014).

This research aims to investigate the applications of artificial intelligence and knowledge representations in management of engineering megaprojects, with the specific focus on estimating CapEx contingencies. It primarily uses linked data, semantic web and ontology engineering to create a knowledge base and BBNs to create intelligent decision making model. Therefore, it aims to build on the combination of logical deductions and probabilistic inferences of the two methods to systematically solve this problem (Zangeneh et al. 2017).

References

Abdelgawad, M., and Fayek, A. R. (2010). Risk Management in the Construction Industry Using Combined Fuzzy FMEA and Fuzzy AHP. Journal of Construction Engineering and Management, 136(9), 1028–1036.

An, S., Kim, G., and Kang, K. I. (2007). A case-based reasoning cost estimating model using experience by analytic hierarchy process. Building and Environment, 42(7), 2573–2579.

Berners-Lee, T., Hendler, J., and Lassila, O. (2001). The Semantic Web. Scientific American, 284(5), 34-43.

Emsley, M. W., Lowe, D. J., Duff, A. R., and Harding, A. (2002). Data modelling and the application of a neural network approach to the prediction of total construction costs. Construction Management & Economics, 20(6), 465–472.

Flyvbjerg, B. (2014). What you should know about megaprojects and why: An overview. Project Management Journal, 45(3), 6–19.

Flyvbjerg, B., Holm, M., and Buhl, S. (2002). Underestimating costs in public works projects: Error or lie? Journal of the American Planning.

Gypton, C. (2002). How have we done? Engineering and Mining Journal.

Hickson, R. J., and Owen, T. L. (2015). Project management for mining : handbook for delivering project success. SME Press.

Hollmann, J. K. (2016). Project risk quantification : a practitioner’s guide to realistic cost and schedule risk management (1st ed.). Probabilistic Publishing.

Khodakarami, V., & Abdi, A. (2014). Project cost risk analysis: A Bayesian networks approach for modeling dependencies between cost items. International Journal of Project Management, 32(7),

Koller, D., & Friedman, N. (2009). Probabilistic Graphical Models. Cambridge, Massachusetts: MIT Press.

Lowe, D. J., Emsley, M. W., and Harding, A. (2006). Predicting construction cost using multiple regression techniques. Journal of Construction Engineering and Management, 132(7), 750–758.

Merrow, E. W. (2011). Industrial megaprojects: concepts, strategies, and practices for success. Wiley.

Nasir, D., McCabe, B., & Hartono, L. (2003). Evaluating Risk in Construction–Schedule Model (ERIC–S): Construction Schedule Risk Model. Journal of Construction Engineering and Management, 129(5), 518–527.

Pauwels, P., Zhang, S., & Lee, Y.-C. C. (2017). Semantic web technologies in AEC industry: A literature overview. Automation in Construction, 73, 145–165.

Zangeneh, P., McCabe, B., Pearson, M., and Mason, N. (2017). Representation and management of project’s knowledge – a linked data approach. Proceedings of the 6th CSCE/CRC International Construction Specialty Conference. Vancouver, Canada, May 31st – June 3rd 2017. (Submitted)

Singleton Jr., R. A., Straits, B. C., & Straits, M. M. (1993). Approaches to social research, 2nd ed. (5th ed.). Oxford University Press.