Risk analysis in the evaluation of plant investments: The contribution of a non-deterministic...

In every human activity, there is a certain margin of risk: each time a decision is made with effects that will appear in the future, it must be taken into account that the hypotheses, based on which the various alternatives are chosen, are never certain but only statistically probable (Associazione Italiana Commercio Estero, 1995). Thus, instead of an absolutely certain result, there will be, in general, possible solutions, and each has its own given probability of occurring.

Therefore, in many activities, it becomes of paramount importance to identify the sources of risk and to understand how these will affect the final result: In fact, risk analysis is proposed to deal with the problems involving uncertainty by identifying, evaluating, and monitoring the risks.

The risk manager's main task is to analyze a more or less complex process in order to identify and quantify the nature of the risk involved, evaluating the possible consequences associated with it: Events with a low probability of occurring might generate just as unfavorable and problematic consequences.

The use of risk analysis techniques has led to good results in some industrial sectors. As far as economic parameters are concerned, this is especially the case for the engineering company sector in which the occurrence of negative events, that are difficult to predict during the project estimation phase, has often led to noncompliance with the predicted cost and time targets. Forecasting is a difficult process due to the effect of classic estimation factors, but also because of other factors, such as the development rate of new technologies (for which references cannot be made relative to experience involving projects that already have been completed), a time shift in the return on investment (that therefore becomes more uncertain), and the greater environmental and safety restrictions.

Thus, engineering companies must focus greater attention on forecasting and management problems, in order to predict the effects of numerous uncertainty factors, while, for those buying the plants, it is of fundamental importance for a project offer to include an evaluation of the reliability of such forecasts.

The risk analysis techniques, along with the ordinary cost and time estimation, provide a measurement of the reliability of the estimate at the moment in which it is made, reducing the probability that, once completed, the estimated values will be exceeded and, as a consequence, avoiding a reduction in the overall profitability of the entire project.

In terms of methodology, risk analysis can be divided into three consecutive stages:

* The first stage involves an analysis of the data with creation of the probability distribution for the variables involved;

* The second stage concerns the propagation of the probability distributions for the variables involved to obtain the required output distributions (this can be carried out using the Monte Carlo simulation or the discrete probability distribution method);

* The third stage requires an evaluation of the uncertainty impact on the project estimated results; in this way the decision-making process tends to be supported by quantitative data and not based only on personal evaluations.

Due to problems in interpreting the results, usually there are significant interactions between risk analysts and decisionmakers (Covello, Lave, Moghissi, & Uppuluri, 1987).

In many situations, an analytical model may be adequate to guide the decision-makers toward the best choice (Wiston, 1996). The problem is that, in some cases, analytical solutions do not exist. In fact, in other uncertainty situations, it is difficult or impossible to construct an analytical model that can provide useful information for decision-makers. In these cases, a simulation may be used that can be generally defined as the technique used to build risk models that can imitate a real situation.

For example, the simulation can be used to determine the sensitivity of a system to the variation in some operating conditions. Let us consider the case involving a company that must choose a single investment from among several possible projects. If the future cash flows of each investment are known with certainty, then most companies will choose the one with the highest net present value (NPV). On the other hand, without certain data, the decision will be more difficult.

By using simulation-based risk analysis models it is possible to obtain a frequency distribution for the NPV of the project.

This makes it simpler to answer questions such as:

* What is the riskiest project?

* What is the probability that the NPV of the investment will be less than or equal to a given value?

* What are the estimated mean and variance of the project NPV?

In this way, it becomes easy to answer these questions, even for different future foreseeable scenarios.

Thus, in general, constructing a risk analysis model in the planning stage helps to evaluate the future prospects without certain data and, thus, makes it possible to evaluate an investment while reducing to a minimum the risk of exceeding the estimated costs in the future.

From what is strictly a modeling viewpoint, with the risk analysis we shift from acting in a situation of certainty to acting in a situation of uncertainty to offer decision-makers a range of possible target behaviors based on scenarios that are probabilistically known or hypothetical for the independent variables.

Application to a Real Case

The proposed study concerns the analysis of the economic feasibility of the project involving a chemical additive production plant in the area of Jobail in Saudi Arabia. The aforementioned plant consists of four independent production lines that produce anti-oxidizing, light-stabilizing chemical additives.

During the economic financial feasibility study of the project, two important results emerged:

* An estimation report indicating the resources used, the relative cost figures, cost trend, and a forecast for the operating profits;

* An analysis of the risks related to the variability of the most significant parameters and, thus, a probabilistic forecast of the various scenarios that may occur.

To make the general situation regarding the prospects as clear as possible for the potential investors, figures were submitted that were obtained using standard estimation techniques and a profitability analysis applying a risk analysis model.