Statistics are like bikinis. What they reveal is suggestive, but what they conceal is vital Aaron Levenstein

The March issue of the Journal presents an exceptional set of prestigious papers that had been presented at conferences in Chile, Australia and South Africa on long-range strategic planning. It was difficult to pick out any one paper for comment and of course the topic is so well covered in the presentations and in a vast amount of literature that to attempt to condense it into a short Journal Comment would be difficult. However, one particular paper caught my eye as a topic of personal interest and could also be related to the next issue of the Journal, which contains a selection of papers from student research projects. Long-range strategic planning in relation to research and technology is not often a topic in conferences and journal contributions, and it was for this reason that one of the papers in the March issue fascinated me.

This is ‘Modelling financial risk in open pit mine projects: implications for strategic decision-making’ by authors from Canada. They provide in some detail the methodology used for the future planning of a proposed open pit mining proposition. They compare the standard analysis, which they refer to as the ‘Static DCF cash flow analysis’, with the ROV analysis (Real Option Valuation). Both methods take into account the wide range of risk factors, such as future variations in prices and markets of the products, in an attempt to determine the proposed life of the open pit mine. This is of course a primary strategic decision needed to proceed with the design of the mining plan.
They selected the well-known Monte Carlo method to take into account the wide range of risk factors to calculate the NPV, NCF and optimum life of the mine. The authors describe in much appreciated detail the methods used to calculate the conventional cash flow analysis giving the NPV, NCF and the DCF and ROI. This is referred to as the Static DCF. A second set of values is calculated called the ROV. The difference between the Static DCF and the ROV analysis is that the latter method takes into account, on a statistical basis, that the management has the option to terminate the mining operations if the risk factors change unfavourably. The startling results of the calculations are that the Static DCF predictions are to plan for a 15 year life whereas the ROV methodology indicates that a 20 year life of the open pit could be the optimum choice. At first sight, the ROV result seems to offend common sense interpretation of the DCF result.
As is well explained in the paper, this relates to the option of building the process plant over a peripheral part of the deposit, which will preclude the opportunity of achieving positive cash flow in later periods. With this aspect the ROV conclusion makes financial and strategic sense. It was no surprise to me since in the early 1990s, I presented a paper at the last of the Commonwealth mining and metallurgical conferences in South Africa, in which I came to the conclusion that, if I include abort or revise options in the statistical analysis of research projects, I could improve returns on investment considerably.
My analyses were not as rigorous as those in this paper—and I was pleased to note the more sophisticated methods, which have emboldened me to propose some South African real options for evaluation as follows:  ‘Picking the eyes’ out of a deposit compared with ensuring that low grade sections can be exploited at a future date  Leaving low grade coal in an open pit compared with removing all coal, the low grade portions to be utilized at a future date  Treating acid mine drainage by adding lime and dumping sludge of precipitated toxic metals on a slimes dam as compared with a total recovery of all metals and water to ensure a zero waste strategy  In treating low grade uranium ores in desert regions by acid leaching and anion exchange techniques can one quantify the option of recovering all materials dissolved from the barren solutions to enable the water to be recovered and reused or used to grow crops? This to be compared with the evaporation of these waste solutions on slimes dams.
I am sure that the answer is positive for the ROV method provided we can allocate statistical determinants to the various options. On the basis of my previous paper I believe this can be done and one hopes this might lead us to some valuable advances in strategic planning with implications in venture capital evaluation. But what has such lofty hypothetical thinking to do with students’ research projects? It is my perception that, at our universities, the academics and advanced students and postgraduates love computer modelling and its applications, especially in mining economics, with lots of log normal distributions and loads of fuzzy logic with which to play with. The standard static DCF methodology is commonplace in all business schools and basic to all bankable feasibility submissions. I am not sure to what extent the ROV approach has penetrated as a standard tool.
The authors use the Monte Carlo statistical methods for the numerical analysis and point out that there are other methods. The extrapolations used as examples require more than a standard statistical protocol. The technical aspects must be put into a number of cash flow conceptual analyses and this could be done in the same hypothetical model used in this paper. These could be wonderful exercises for final year student projects and/or postgraduate theses at an MSc level. Obviously this would be done with the collaboration of staff and students with industry and statutory bodies such as Mintek.
This paper from Canada is excellently referenced and could well serve as a primer to introduce final year students and postgraduates to the methodology and the looked-for outcomes in these examples and countless other strategic opportunities. It would also introduce staff and students to the basics of strategic planning, and I maintain that this is now becoming a crucial part of technological development in any country. There are aspects of strategic planning that have to be integrated with the various fields of technical advancement and cannot be left to the business schools lacking the latest technical challenges. But such evaluation methods are becoming essential components of science and engineering education. Maybe these students’ projects on strategic evaluation methods could evolve into an ‘MTE’ postgraduate school (Master of Technology Evaluation), to be established in the Engineering faculties with close interaction with graduate course syllabi and postgraduate studies. This paper has revealed some enticing suggestions and maybe could lead to some vital aspects demanding attention to Gaussian curves and even fuzzy logic.  R.E. Robinson March 2009