Principal Investigator Gregory McRae
Much of the recent research on risk-management has arisen out of Professor McRae's participation in the BP Projects Academy operated by the Sloan School and School of Engineering in association with BP's top management -- where part of the course involves themes of understanding, managing and mitigating financial, commercial, country and technical risks.
McRae's research has focused primarily on how to deal with risk in project management by developing tools that help decide where to spend money most effectively to reduce uncertainty. This research is very useful in R&D and capital investment portfolio management. The models work by propagating uncertainy through complex systems and processes, which can model chemical processes, R&D processes, capital investment and development processes, and supply chains.
Examples of this type of challenge are:
(1) in the pharmaceutical industry, when facing the opportunities to invest large dollar amounts in a diversity of drug candidates for development, understanding the implications of investments in each candidate. Which candidates carry low vs high risk? What is the value of collecting additional information for each candidate? One can put a lot of investment in the front-end of a project (initial R&D expenses) but the impact on reducing risk may be different for different candidates.
(2) in the oil industry, when prospecting for new reservoirs, how much do you invest in any particular location when you have multiple potential production fields that could be developed, each with its own risk profile, and an overall production and cost objective to be met?
The modeling tools additionally have an impact in financial portfolio management, receiving additional validation last summer when McRae spent the summer working with Goldman Sachs applying them to the evaluation of financial risks. The Black-Scholes models traditionally applied to calculate financial risk in pricing options, etc., are based upon stochastic/Brownian processes, but time series studies have shown big jumps in prices that are not modeled effectively by these equations. Newer models predict these jumps, and have much the same structure as the models with which McRae has been working for risk management in other topic areas.