Getting the most from economic modelling
Feature Articles, Mar  16  2009 (Digital Energy Journal)

- The use of complex economic models is sharply increasing in the oil and gas industry, says Bart Willigers, senior consultant with oil and gas financial modelling company Palantir Solutions. Here are some tips for making the most out of it.

Bart Willigers, senior consultant with oil and gas financial modelling company Palantir Solutions

“Oil and gas companies have a poor track record at delivering the results they promised as budgets over run and time lines are stretched,” says Bart Willigers, senior consultant with oil and gas financial modelling company Palantir Solutions.

The problem is that the decision systems the industry is using aren’t good enough to make a reasonable prediction of what is going to happen in the future.

“Having an efficient decision making system in place is as important as having very bright geoscientists and engineers,” he says.

Most oil and gas companies employ individuals who are responsible for both the financial reporting on past events and the economic modelling of ongoing and future activities – the problem with this set up is that the two tasks require totally different skills and mind-sets.

Meanwhile many industries, including pharmaceuticals and electronics, treat decision analysis as a discipline in its own right and employ specialists who can help people make the right decisions using a range of different tools.

However, the oil and gas industry has started to appreciate the value of decision analysis and the decision models used are getting more and more sophisticated. “A few years ago – they would build one scenario – there’s the price you’re going to use – and they didn’t look at it again,” he says. “Now there are companies making much more complex models, looking at different prices and different development opportunities.”

The profitability of a project is a function of a number of uncertain variables: the volume of hydrocarbons in the ground which can be produced, the oil price and the gas price, rig rates, steel prices, concrete prices and labour.

Taking a prediction of each of these variables, computer technology is used to determine the likely outcome of the entire project - the amount of profit it is 90 per cent likely to achieve, and what (upper and lower) profit it is 10 per cent likely to achieve.

The application of state-of-the-art computer programs, like the suite of software tools developed by Palantir Solutions, enable the industry to get much better at hitting its targets, he says.

Decision tools are very helpful in trying to get objective answers. “If you ask someone – how likely do you think it is that we find oil in this well – people will tell you different things,” he says. “Someone might have a different opinion – or a different degree of risk aversion. There can also be differing opinions within the project team, about whether the objective is to have the field online as soon as possible, or to add reserves over the longer term. We have to remove this bias.” he says.

Scaling it right

For the model to be useful, it is important that it does not include more variables than it needs.

“You can usually define plenty of uncertainties. But if you do too many – you end up with a monster model which is unworkable,” he said. “You have to boil down these uncertainties to a number you can manage. You start with 20 and end up with four.”

“You can make these models as complicated as you want, but you typically don’t get more insight by making things overcomplex,” he says. “You’re trying to filter out all the noise – so have the drivers which lead to the success of the project. At the same time you need to be able to show why you have included these things and not others.”

Palantir Solutions

Palantir Solutions offers economic modelling services, software and training to the oil and gas industry. It has offices in Aberdeen, London, Houston, Calgary and Singapore, with software development in India.

It organises courses to help people work out how to model risk and communicate their ideas to the project team, and work out what information they need to build a model.

It aims to understand how different input variables might affect each other – for example, a reduction in oil price is likely to be followed by a reduction in rig rates. “We’re trying to model the relationship between prices – try to capture the dynamics you see in the marketplace,” he says.

Palantir Solutions is also working out ways to remove ‘outlier’ or ‘wacky’ numbers from the MonteCarlo simulation results –outcomes which look so far outlandish they could never happen. Or could they?

One tool which is uniquely capable to rapidly go through large amounts of data and screen it is a product developed by Spotfire, a company that is working with Palantir Solutions “For example, you might have five per cent of the results in a far corner, far away from the other data, which are less likely to be overlooked when displayed graphically”, he says.

Mr Willigers has a Phd in Geology and an MBA, so he is uniquely skilled to help oil and gas companies make business decisions based on geologic data.

“I basically work on site with clients on the economics – assisting with the planning process, figuring out what to do with different assets,” he says.

A big challenge is integrating companies’ IT systems together. “What we’re trying to establish – is a series of products completely linked together – so you don’t have to copy paste data from one system to another, keeping track of hundreds of excel files,” he says.

Palantir Solutions

<<BACK