Evaluating and Managing Infrastructure Mega-Projects in Developing Countries: A Novel Approach
Artem Khudenko, University Of Liverpool, P. R. Drake, University Of Liverpool
To tackle the afore-mentioned issues, we propose a new simulation-optimisation holistic approach to appraising and managing infrastructure mega-projects.
Large-scale infrastructure investments, considered by many scholars as a very important component of revitalising the crisis-hit economies, often fail to bring the expected returns. One of the reasons behind this may be that the government, required to optimally ration nation’s scarce resources amongst competing opportunities, usually acts irrationally by strongly pushing mega-projects with overabundant optimism or vested interest, notwithstanding their disastrous performance history. In emerging economies, the instance is accentuated by the volatile markets and largely unpredictable government’s macroeconomic course. To guarantee public capital productivity maximisation, authorities and project managers are now practically obliged to use some form of a sophisticated decision support system. However, our research shows the abundance of flaws in and overall scarcity of dedicated decision making tools that would support infrastructure strategists and managers in developing countries.
In theory and practice, both in the developed and developing worlds, infrastructure investment project appraisal is predominantly facilitated by the Cost-Benefit Analysis (CBA) and Multi-Criteria Analysis (MCA). However, these frameworks lack in capturing system dynamics, uncertainty, and managerial decision flexibility, being also awkward in evaluating cohesive infrastructure networks rather than isolated projects.
To tackle the afore-mentioned issues, we propose a new simulation-optimisation holistic approach to appraising and managing infrastructure mega-projects. To this end, we have developed a dynamic model of the transportation system, based on Monte Carlo simulations of crucial uncertain variables, expressed as modified Wiener continuous stochastic processes with jumps. We analytically formulate and incorporate an optimisation algorithm for the sequencing of interdependent infrastructure elements. In order to explicitly allow for decision flexibility at each project stage, the Real Options theory is utilised. On the demand side, the gravity and mode choice models are employed to elicit the expectations on infrastructure usage. We also attempt to capture transport network effects by broadening the spatial boundaries of the conventional project evaluation scope. In this regard, the model facilitates both a nation-wide network and transnational integration setting and offers a rational degree of network expansion in terms of both the return on investments and estimated level of project risk.
We then test the proposed methodology with the cases of high-speed rail development in Eastern Europe (Ukraine and Russia) within the established integrative framework of Trans-European Transport Network. Compared to the conventional, static and deterministic, infrastructure appraisal methods such as CBA and MCA, the reported stochastic dynamic model appears suitable not only for ex ante project evaluation, providing a strict feasibility verdict for now, but also for being employed amidst project implementation, emphasising conditions of when it is best to initiate, defer or discard a certain infrastructure construction stage.
Such a holistic tool is envisaged to enable decision and policy makers in emerging economies to evaluate and adequately manage infrastructure mega-projects, whilst examining possible intra- and international expansion strategies based on a trade-off between their financial performance and quantified risk.
Association for European Transport