There is some scepticism about the value of critical path analysis techniques utilised in delay expert evidence.
I suspect that a key aspect of the sceptics’ problem arises from the application of prospective methods of delay analysis that predict the effect of delays at any given point as compared to a purely fact based approach that analyses what has actually happened. In particular, difficulty arises where prospective methods of analysis are applied retrospectively after project completion with results that do not accord with the facts.
I have no doubt that delay analysts can point to plenty of examples where the use of a programme presentation utilising one or more of the commonly used delay analysis techniques has played an important if not determinative part in the outcome of the assessment of a claim for delay. There are a growing number of court cases where this has been so.
But equally there are examples of disputes in which huge amounts of work have been undertaken in the production of complex analyses that have been of little or no value and ultimately have been entirely rejected by the tribunal. The judgments are frequently critical of one or other expert analysis and very occasionally both, but this is often about deciding the merits of the case and which evidence is to be preferred and the justification for that decision. This is often about the approach adopted by the expert and the extent to which he demonstrates thoroughness, independence and credibility. It is my view as both a delay analyst and an adjudicator that delay analysis expert evidence can be of considerable assistance if it is approached in the right way.
The starting point for the debate is the nature of the obligation under many standard forms of construction contract for the architect or engineer to decide whether an event has caused or is likely to cause a delay to completion and then to apply some degree of discretion in deciding such extension of time that is considered to be fair and reasonable. Further, although the contractor is under an obligation to give notice of a delay and provide particulars to support the contention that the event has or is likely to affect the completion date, the JCT and ICE forms provide an obligation on the certifier to undertake a final review of all matters it is aware of regardless of whether the events have previously been notified by the contractor.
There is likely to be a significant difference in approach to analysing the effect of delay depending on whether the project is in progress at the time of determining the entitlement or whether it is complete. In the former case the ultimate effect of the events claimed as causing delay can only be estimated at the time of the determination, and some sort of prospective method of delay analysis has to be relied upon. This is the approach required under the NEC contracts for example. In the latter case it is known when the project was complete and it is then possible to look back at what has actually occurred. Whilst a dispute about a delay may be referred to adjudication before the completion of the works, requiring some degree of prospective approach to modelling the effect of the delay, in the main a dispute about delay that comes before the court will arise out of a project that has been completed.
The point is well illustrated by the Hong Kong case of Leighton Contractors (Asia) Ltd -v- Stelux Holdings Ltd (2004), in which Leighton argued that the contract made it clear that both “delay” and “likely delay” gave proper grounds for an extension of time. Leighton contended that by standing in the architect’s shoes at the time of the delay events, if the arbitrator thought that an event was likely to cause delay, she should have granted an extension of time. It was irrelevant that, with hindsight, the event did not in fact delay completion. The court rejected those contentions and upheld the decision of the arbitrator. The court confirmed that though a time slice approach to examining extension of time entitlement may be appropriate, an overly theoretical or artificial result which has no regard to the “as-built” situation will be rejected.
There have still not been a very significant number of court cases involving delay analysis but there are a few essential elements that in my view have emerged from the judgments:
- Delay is a question of fact and, after the event at least, the courts are interested in what has actually happened rather than in what could have happened.
- For an event to affect the completion date it must fall on the critical path of the project.
- Notwithstanding the discretionary nature of the architect or engineer’s duty to award time it has been found that the analysis of delay by a certifier cannot be impressionistic, but should be based on a calculated approach.
- The delay expert must consider all the relevant facts and evidence that have a positive or negative impact upon the opinions expressed in relation to the issues in question.
As to point 1 above, the fact that delay is a question of fact is an argument the delay analysis sceptics say relegates critical path delay analysis to a secondary or supporting evidential role. This I think depends on what type of analysis or work is being referred to by the sceptics.
In circumstances where the project is complete I agree that delay is predominantly a question of fact. In my experience there are several stages to the forensic analysis of delay to a completed project. The first is to ask what should have happened. It may be convenient to agree that this is represented by the contractor’s initial programme, but depending on the quality of this programme, this may not always be appropriate. Then, at high level to start with, the as-built programme is established and compared to the planned programme. There are a variety of possibilities at this stage that may enable a focused approach to be adopted thereafter, for example only certain parts of the work may have been delayed at certain stages within the project. This comparison is then a valuable reference tool. A simple comparison of planned and actual does not tell you why a delay has occurred but it is a good start.
The second step in providing a focussed approach to the analysis of delay to a completed project is to identify the as-built critical path to enable the factual research and effort to be focussed on the events causing critical delay. The third step in relation to each event that is likely to have affected the critical path is to establish the facts concerning the delay itself, its cause, the duration of the delay, the chronology of events that led to the delay and the activities that were directly affected. These are all matters of fact that with reasonable records or statements from witnesses should be capable of being established. It will also be necessary to consider any other facts relating to events that could bring into question whether the events relied upon are in fact critical. It may also be necessary to establish the facts in relation to delay events that are not critical in order to consider the potential costs or loss and expense that arise from such events.
The second step described above is essentially a programme based exercise and a programming function, since it requires experience of construction planning and sequences of operations and therefore a combination of programme analysis and opinion to establish what was likely to have been the critical path.
The third step is to establish the effect of the delay on subsequent events and ultimately on the completion date. It has to be accepted that the critical path might be blindingly obvious, so little if any real analysis may be required, but it still requires evidence to support what the critical path actually is and in my experience it is often of great value even in the simplest cases of delay to set out the chronology and the delay on a simple programme and place it in the context of the rest of the programme, showing how it has affected the critical path.
When considering issues of concurrency it is necessary to establish the relative potency of the potentially concurrent events. This more often than not is a question of both fact and experience based opinion.
As to point 2, it is obvious that the claimant must prove and the tribunal must decide that an event has affected the critical path, which generally means that expert evidence must be adduced as what is the critical path. This is clear from the case of Motherwell Bridge Construction Ltd -v- Micafil Vakuumtechnik (2002).  In the more recent case of Mirant Asia-Pacific Construction (Hong Kong) Ltd v Ove Arup & Partners International Ltd (2007), Judge Toulmin clearly understood that working with critical path analyses on complex projects is not an exact science and that the question of whether an event has delayed the project is always a question of fact. However, without such analysis undertaken by suitably experienced experts, he commented that the parties may be mistaken as to what is on the critical path. He referred to time impact analysis as an excellent method of analysis, but he emphasised that the analysis will only be valid if it is comprehensive and takes account of all activities.
As to point 3 above, in John Barker Construction Ltd v London Portman Hotel Ltd (1996) it was said that in exercising his duty under clause 25 the architect or contract administrator must undertake a logical analysis in a methodological way of the impact of the relevant events on the contractor’s programme. The application of an impressionistic rather than a calculated and rational assessment is not sufficient.
The John Barker Case was later considered in the case of Balfour Beatty Construction Limited v The Mayor and Burgess of the London Borough of Lambeth (2002) in which it was said by His Honour Judge Humphrey Lloyd QC that: “In the context of a dispute about the time for completion a logical analysis includes the logic required for in the establishment of a CPN [critical path network].”
The production of a construction programme is a requirement of most standard forms of contract. The NEC contracts make the programme a central management tool for the project. Even Rev 2 of the JCT contracts now include the option to insist that the contractor shows the critical path on its planned programme. The programme and the effective management of it are therefore becoming an essential part of effective project management. The management process provides a certain amount of data about the status and progress of the works at various measurement intervals. This usually provides important and valuable data about what has happened on a project.
The fourth point refers to the guidance from the cases and the Civil Procedure Rules (CPR) requiring the delay expert to consider all the relevant facts that might bear on the opinions expressed in relation to the issues in question. This requirement suggests the need for what could amount to a time consuming and expensive detailed investigation. The CPR is also however concerned with ensuring as far as possible that the whole process including the input of experts is proportionate to the dispute to be decided. There is obviously a tension between these two imperatives. The lesson from both the Mirant case and the more recent case of Costain Limited v Charles Haswell and Partners Limited  appears to be that whilst it may be preferable for the parties and their experts to identify the core issues and focus on those, there is a clear requirement to consider the whole picture, at least to the extent required to prove the ultimate effect of an event.
In the Costain case the experts both agreed that time impact analysis was the appropriate approach to delay analysis (although they applied it differently) and they both agreed that the event was on the critical path at the time of the delay, but presumably in the interests of proportionality, they focussed their investigation only on the period of delay in question without having any regard for what happened thereafter.
The Judge found that both experts had not undertaken any analysis of the project for the period after January 2003. He said that as a result they had not considered what effect the foundation delays actually had on follow-on activities and other structures elsewhere within the project. Furthermore, the experts had also not investigated if Costain had been able to mitigate or negate (or even exacerbate) the initial delays by later events. He rejected Costain’s claim for the effect of the delay observing that it was “simply a matter of speculation” on Costain’s part.
This might seem somewhat harsh given that both experts had agreed that the events affected the critical path at the time, but the point is clearly that it must be established that the result of an analysis of delay within a particular window of time actually formed part of the ultimately critical delay to the completion of the project. As also illustrated by the Leighton Contractors case referred to earlier, if the delay projected by a prospective delay analysis technique does not accord with what actually eventually happened then it is likely to be rejected. This does not mean that the use of prospective time slice analysis is not appropriate for the retrospective analysis of delay to completed projects. Indeed it was specifically approved of by the judge in the Mirant and other cases. The problem seems to be in the way the technique is applied.
In conclusion, it is clear that fact is king, and regardless of what method of analysis is used, if project delay is analysed retrospectively the result of any analysis must accord with what actually happened. A consideration of the as-built programme may often be sufficient and in my experience there is considerable advantage to marshalling and presenting the as-built facts in the form of an as-built programme, and in identifying the actual delays with reference to it. This alone though may not be sufficient to identify the critical path and other forms of analysis may be appropriate depending on the case and the nature and extent of evidential material available.