Nothing if not critical

Inconveniently, two of the main foundations in a whopping great boiler house settled 46mm and 66mm while steelwork was being erected on them. It all had to be dismantled. You can imagine what might be said about consequent delays.

Hands were held out for £50m in compensation, if you please. They were the hands of Mirant Asia Pacific Construction. These hands had fingers, and they were eventually pointed at the Hong Kong office of Arup. The allegation was breach of contract and negligence.

Whatever the truth of those allegations, said Arup, the trouble with the foundations was not responsible for the five-month delay to the project, which was a coal-fired power station in Sual, in the Philippines. “But the boiler was on the critical path,” said Mirant.

Three High Court trials and six years of litigation later, we come to the fascinating judgment of Judge Toulmin. Before I tell you the story, I should acknowledge my debt to one of the key programming experts in the trial: Mr Ian Robinson, a partner in Davis Langdon. He was praised by the judge for his delay analysis.

Credit: Simone Lia

"Ah! don't bother with that computer programme - i know where the critical path lays. I can feel it in my bones."

The difficulty was that the case was largely about how to measure and assess the effects of competing causes of delay, and in particular how to do so in cases where critical path analysis is difficult because of shortcomings in a project programme. In other words, the usual.

Arup argued that there were numerous causes of delay in addition to the foundations. Mirant responded that without that bother, all the other delays could have been recovered. Arup’s expert did not agree; he said all the other delays would have resulted in loss of several months. So, the dilemma for the tribunal, was to decide which delays were dominant. Mind you, it would have helped somewhat if reliable programmes had been on hand. There were none. Witnesses came along all convinced that they knew where the critical path lay. It turned out that they were working on little more than intuition. Oh ho, so what else is new?

Witnesses were convinced that they knew where the critical path lay. It turned out that they were working on little more than intuition. Oh ho, so what else is new?

The judge wasn’t happy to decide a case on instinct and guesswork. The witnesses might have been wrong about what they thought was the critical path; the judge himself was unable to reach a firm conclusion as to whether the foundations were on the path.

Judge Toulmin thought it was necessary to look at all activities that might have been on or close to the path. It is an error for a programming expert to confine his or her analysis to those activities that lie directly on the path while ignoring the effect of those that may have a minimal amount of float but are nevertheless close enough to it to influence the outcome.

The expert had even given the court a health warning about the unreliability of the programmes and went on to carry out a review of the contemporaneous documents. The judgment shows just how reliant tribunals are on experts. And let me throw in my twopennyworth here. The tribunal will look for corroborating information to show that the expert isn’t pulling a fast one. The judge could see “significant clues” in the documents to support the critical analysis.

By the way, all you experts, do play it straight. Strain at the arguments, extend credibility to breaking point and the tribunal will cast you out. You owe your first duty to the arbitrator, adjudicator or judge.

I drew the distinct impression that almost all those involved in the project thought that the boiler house foundations were a critical activity. But in the run-up to the litigation it dawned on the Arup team that the project’s programmes were not reliable enough to determine the facts retrospectively. Better to look at all of the available project records.

In the end the judge dismissed Mirant’s claim … not because it had not proved its case. Rather he did so because he was satisfied, on the evidence, that Arup had demonstrated that the dominant delays were caused by factors other than the foundations.

A footnote: Arup’s lawyer, Beale & Co, reports that Arup was awarded its costs. These were upwards of £8m, which is 40% of the firm’s profit for 2006 and £3m more than its profit the previous year. A good argument to win.