“I always tell the truth, even when I lie” – A Brief Guide to Delay Analysis Techniques

“I always tell the truth, even when I lie”
A Brief Guide to Delay Analysis Techniques

Author: Eugene Lenehan

January 2023

INTRODUCTION

Most delay analysis exercises are a blend of facts and fantasy.

The most frequent cause of construction claims relates to project delay. Identifying the causes of delay usually involves some sort of delay analysis. There are many different methods of delay analysis. Whilst these are often highly technical exercises, generally carried out using sophisticated computer software, many consider the process to be a ‘dark art’. Although there is a perception that use of specialised programming software allows delays to be analysed scientifically, producing precise and reliable results, there is also a suspicion that parties manipulate the process to achieve desired outcomes.

I always tell the truth, even when I lie” is a line delivered by Al Pacino’s mesmerising character (Tony Montana) in ‘Scarface’, a 1983 film directed by Brian De Palma and written by Oliver Stone. It tells the story of a Cuban refugee (Tony Montana) who arrives penniless in the USA and goes on to become a powerful drug lord. The meaning behind Tony Montana’s contradictory statement is that he makes his own truth! I believe that the way disputing parties approach delay analysis is often reminiscent of this sentiment!

On initial review, delay analysis exercises frequently appear plausible, and are invariably based on an assortment of demonstratable facts. However, under closer scrutiny one might find that the exercise is based on unrealistic assumptions, and/or that it neglects to take account of various salient facts, with the inevitable result that the exercise fails to reach the correct conclusions about the actual causes of delay. These problems can be exacerbated by the method of delay analysis.

In Tony Montana’s world, things are however he says they are! But in the world of construction disputes, Judges (and adjudicators) can get to decide whether a delay analysis exercise is correct.

For example, if an ‘Impacted As-Planned’ approach has been used, perhaps it is based on the original, but flawed, programme. Or perhaps the delay exercise has neglected to factor in several contractor culpable delays. Alternatively, if a ‘Collapsed As-Built’ exercise has been presented, accurately recording the start and finish dates of all activities, perhaps it has also incorporated some unjustifiable logic links that generate a bogus completion date.

This article now explains the most popular delay analysis techniques, before referring to various well-known court cases where judges criticised delay experts and their methods.

METHODS OF DELAY ANALYSIS

To demonstrate the cause and effect of alleged delays, this usually involves carrying out some form of delay analysis.

There is no standard method of delay analysis. The law does not dictate the use of any particular technique for proving a delay, although the form of contract might, such as the NEC contract. For most forms of contract, including JCT contracts, it is a matter for each party as to how it presents its case.

Some of the methods are theoretical based methods (such as ‘Impacted as-planned’ – see further below). They show the theoretical impact of delaying events rather than what actually occurred. They do this by theoretical analysis of what the effect ought to have been and not by identifying its actual impact from recorded as-built performance. Given that they are theoretical methods, there is an element of subjectivity.

In The Royal Brompton Hospital NHS Trust v Frederick A Hammond & Ors [2000] EWHC Technology 39 it was also said that “the accuracy of any of the methods in common use critically depends upon the quality of the information upon which the assessment exercise was based”.

The Delay and Disruption Protocol

In 2017 the Society of Construction Law published its Second Edition of the ‘Delay and Disruption Protocol’ (the “Protocol”). It provides a best practice guide for the management of extensions of time and disruption entitlements.

The Protocol includes six different methods of delay analysis. Some of the methods of analysis are relatively crude, whilst others can be very complex, time consuming and expensive to prepare.

The six methods explained in the Protocol are as follows:

  • Impacted as-planned analysis.
  • Time impact analysis.
  • Time slice windows analysis.
  • As-planned v as-built windows analysis.
  • Longest path analysis.
  • Collapsed as-built analysis.

These are not the only methods that exist, and they each have secondary derivative methodologies.

Several of the most well-known methods, including most of the methods mentioned in the Protocol, are explained below.

Prospective v Retrospective

Delay analysis methodologies are commonly categorised as either ‘prospective’ or ‘retrospective’, which represents the point in time the analysis is performed.

The word ‘prospective’ is defined as ‘likely to happen’. Accordingly, prospective methodologies (such as ‘time impact’ or ‘impacted as-planned’ analyses) involve assessing delay before they occur, or contemporaneously i.e., at the time the events in question occurred. By contrast, the word ‘retrospective’ is defined as ‘something that happened in the past’. Hence, retrospective methodologies (such as collapsed as-built) take into account subsequent events and attempt to establish whether the events in question actually caused delay to completion.

Sometimes the debate over ‘prospective vs retrospective’ is referred to as a ‘forecast vs actual” issue.

In Fluor v Shanghai Zhenhua Heavy Industry Ltd [2018] EWHC 1 (TCC), Edwards-Stuart J stated:

There has been an extensive debate about the correct approach to delay analysis. … A prospective analysis … does not necessarily produce the same answer as an analysis carried out retrospectively.

As-planned versus As-built

This is perhaps the simplest method of delay analysis. It is a retrospective technique, in which the contractor compares what actually happened to what was planned to happen. The comparison is generally made in respect of the activities on the critical path. Hence, this method requires the as-built programme, which is not always available.

This method does not rely on logic-based programmes or software packages, but is most applicable when the as-planned critical path and the as-built critical path are similar.

This method compares planned start and finish dates with the actual start and finish dates of activities on the critical and near-critical paths. The initial aim is to identify delayed starts, extended durations, and late finishes. Explanations are then offered as to what caused the deviations to the planned progress.

In City Inn v Shepherd [2007] the judge said “the weakness” of this method is:

it does not identify the critical path and therefore needs to be used with great care and understanding of the processes in the whole of the project”.

The limitations of this method include as follows:

  • Accuracy will be compromised by deficiencies in the as-planned programme and/or as-built programme.
  • It does not take into account changes in the sequence / critical path. Hence, its accuracy reduces as the actual critical path deviates from the planned schedule.
  • As-built programmes are not always available.
  • Causation is based on opinion.
  • It is not suitable for complicated projects or projects built significantly different than planned.

Impacted As-Planned

This is a simple type of delay analysis, which involves introducing delay events into a logic-linked baseline programme that reflects the original plan for completion of the works. The programme is then recalculated to establish the likely effect of the delay events, including any new completion date. The various delays are formulated as activities and added to the as-planned programme in a chronological order, determining the effect of each individual delay. The amount of critical delay equals the difference in completion dates between the programmes before and after the impacts.

If a good as-planned programme exists, but it has not been regularly updated with progress and there is a lack of as-built records, then ‘impacted as-planned’ analysis may be suitable.

This technique identifies the effect of an event on the completion date assuming everything else went exactly as planned in the baseline programme.

The technique is relatively easy to perform, and does not require as-built information, which means it can be used for both retrospective and prospective analysis i.e., this method can be used for analysis of delay during and after project completion. However, since actual progress is not considered, this approach may not demonstrate what actually caused the delay to the project.

The disadvantages of this method include as follows:

  • It assumes the original planned programme is correct / adequate, when in fact it might be flawed / unrealistic.
  • It does not consider subsequent changes to the original planned intent.
  • It takes no account of what actually occurred, which might include the contractor’s own culpable delay and mitigation measures.
  • The results are theoretical and might produce a completion date later than when completion actually occurred.

Collapsed As-Built

This technique represents the opposite approach to the ‘impacted as-planned’ methodology.

The first edition of the Protocol’ (published in 2002) defines the ‘collapsed As-Built’ method as follows:

A method of delay analysis where the effects of events are ‘subtracted’ from the as built programme to determine what would have occurred but for those events”.

The method starts by creating and then reviewing a logic-linked as built programme. Whereas the impacted as-planned’ technique is based on the as-planned programme, systematically introducing each delay event in chronological order, the ‘collapsed as-built’ method starts with the as-built programme, and then extracts from it the apparent effect of delay events (starting with the latest), to identify what would have happened during the project but for the effects of the delay event.

This is a retrospective delay analysis method, taking into consideration what actually happened during the construction process, as opposed to what might have happened. It is also classed as a form of ‘but for’ exercise. Upon removal of each individual delay event, the resulting programme will give the completion date of the project ‘but for’ the delay.

This method can be very persuasive, as it is based on what actually occurred, and explains what would have happened had the delays not occurred.

Plenty of good, reliable as-built records are essential to enable a logic linked as-built programme to be created.

Disadvantages of this method includes as follows:

  • Constructing the as-built programme is costly and time consuming.
  • It requires good as-built records. The opposing side is likely to challenge the as-built start and finish dates, making contemporaneous documentation critical to establishing the dates.
  • In collapsing the programme, the delay analyst is often required to insert ‘after-the-fact’ logic links, which can be subjective, not necessarily reflecting the thinking of the contractor during actual performance. Hence, it is susceptible to manipulation.

‘Windows’ Analysis

Windows analysis is a forensic technique, where the construction process is viewed in multiple ‘windows’, or slices of time during the period of performance.  The end of one window acts as the start of the next window.

The total project duration is divided into a number time periods (windows), often monthly, based on dates of interim progress reviews, or major project milestones.

This method evaluates as-built performance information to determine whether activity delays occurred and if so, whether they actually delayed the critical path.

There are two types of windows analysis: ‘time slice windows’ analysis and ‘as-planned vs. as-built windows’ analysis. Both methods aim to determine the contemporaneous or actual critical path through the project.

The time slice method requires a logic linked baseline programme, plus updated programmes coinciding with the start and end of each of the windows. These updated programmes can either be contemporaneously issued programmes or an original as-planned programme that is updated (using programme software) with as-built information after the fact.

The programme within each window (i.e., time slice) is updated to reflect the actual durations and sequences at the time of the delay. The analyst determines any changes to the critical path and the extent of delay to completion. Any change in the completion date identifies the amount of delay during that window period. The analyst then must consider what caused the identified delay within each window. This involves an examination of the contemporaneous records associated with the critical path activities.

In the as-planned vs. as-built windows method the critical path is established not by using programming software but instead by a common-sense and practical analysis of the available facts. The extent of delay is established by comparing the as-planned and as-built dates for the activities along the identified critical path.

The following comments on Windows analysis were made in the case of Mirant Asia-Pacific Construction (Hong Kong) Ltd v Ove Arup and Partners Ltd [2007] EWHC 918 (at paras 131 and 132):

Windows analysis is the most accepted method of critical path analysis. As Pickavance makes clear at page 572 of his book, “Windows” (and “Watersheds”) are not methods of analysis in themselves: they are merely aspects of conducting the critical path analysis. In essence they represent the division of the overall construction period into smaller periods into which each new set of corresponding progress can be entered into the programme and analysed

The term “Windows analysis” refers to the regular reviews and updates undertaken by the contractor, normally monthly. These periods of time would be described as monthly windows. Unlike previous monthly reviews, the planner would use sophisticated software programmes to plot which activity or activities were on and which were near to the critical path each month. The programmes would take into account those activities which had started early or had been delayed. Also built into the programmes would be the progress of those activities which had started since the previous monthly window. This would enable the employer and the contractor to analyse over the relatively short periods of time what changes had occurred, and identify what problems needed to be investigated and put right.”.

The Windows methods can provide a very useful approach to analysing delays, with the more windows or time slices used, the better the accuracy of the results. This method has the advantage of dividing a complicated programme into a manageable one. It also can take into account the dynamic nature of the critical path.

However, limitations of the Windows methods include as follows:

  • It can be a time consuming and costly method, requiring detailed project records, which are often unavailable.
  • If periodic progress updates are missing, the delay analyst may need to carry out a highly arduous analysis of project records to create updates.
  • How delays are assessed within each window can be subjective.

Time Impact Analysis

The ‘Time Impact Analysis’ determines the expected effect of an event on the completion date from the point in time when it was instructed or arose.

This method is recommended in the first edition of the Protocol, defining it as follows:

Method of delay analysis where the impacts of particular delays are mapped out at the point in time at which they occur, allowing the discrete effect of individual events to be determined”.

This technique involves introducing delay events into an updated programmes that reflects the status of the works at the point the delay events arose. This method of delay analysis is consistent with the NEC contract’s methodology.

This method is a prospective (i.e., forward-looking) analysis technique and provides a basis for determining the expected effect of an event on the completion date. The first step is to update the (logic-linked) planned programme, with actual progress up to the start of each delay event.  This is usually done contemporaneously. The programme is then impacted with the delay event, noting the likely effect of the delay events on the then-planned completion of the works. The difference between the new completion date and the completion date prior to the exercise indicates the critical delay caused by that delay event.

It is considered an effective and useful technique because the delays are analysed using real time information. The first edition of the Protocol also stated as follows:

The Protocol recommends that this methodology be used wherever the circumstances permit, both for prospective and (where the necessary information is available) retrospective delay analysis.”

However, although the second edition of the Protocol altered its position, recommending a “common sense approach” that is “based upon an appropriate delay analysis”, it nevertheless continued to say that the time impact analysis method “is recommended for a contemporaneous analysis”.

In terms of disadvantages of this technique, the first edition of the Protocol states that “it is generally the most time consuming and costly when performed forensically”. This is in part due to the delay analyst having to perform highly laborious trawl of project records to creates updates. Obviously, it would be much easier if undertaken as the works progress.

Other limitations of this method include as follows:

  • It is dependent upon the reasonableness of the planned programme.
  • It is considered a hypothetical model as it does not rely on as-built data
  • It may not be practical or realistic to use if there are a vast number of delay events.

JUDICIAL CRITICISM

It is generally considered that identifying the correct critical path is the most important aspect of a delay report. However, in the case of Balfour Beatty Construction Ltd v The Mayor & Burgesses of the London Borough of Lambeth [2002] EWHC 597, Judge Humphrey Lloyd said:

From the material available to me it is clear that BB did little or nothing to present its case in a logical or methodical way. Despite the fact that the dispute concerned a multi-million pound refurbishment contract no attempt was made to provide any critical path.”

Judge Humphrey Lloyd explained “all that BB have provided is an aggregate of all the planned time and compared with all the actual time on a trade by trade basis. No attempt has even been made by BB to demonstrate any link between the trades”.

Many delay analysis exercises are based on the original programme, but if this is flawed in some way, it should first be corrected. The judge in Balfour Beatty v Lambeth [2002] said as follows:

the foundation must be the original programme (if capable of justification and substantiation to show its validity and reliability as a contractual starting point)”.

If an original programme does not correct any flaws, it is merely an unrealistic plan and likely to be rejected by the court, as was the case in Skanska Construction UK Ltd v Egger (Barony) Ltd [2004] EWHC 1748, where the judge said as follows:

in order to carry out its impact analysis the master programme prepared by Skanska on 23rd May which for reasons extensively set out in the liability judgment became virtually redundant, almost from the outset because of the late provision of vital information relating to design and layout and changes made on the instructions of Egger. I am satisfied that the sub-zone programmes dealing with each zone separated in planning and programme terms provide a more accurate basis for detailed delay analysis rather than the flawed planned programme”.

As impressive as a sophisticated delay analysis might appear, one must always be wary of exercises based on ‘junk in, junk out’. In the Skanska case, Egger’s expert had “produced a report of some hundreds of pages supported by 240 charts”. However, it was found that the analysis relied on untested facts, that were subsequently proven to be incorrect. Whilst the judge referred to the expert’s opinion as being based on a “sophisticated impact analysis”, he also explained that “however impeccable the logic of that programme” it was “only as good as the data put in”, and “The court cannot have confidence as to the completeness and quality of the input into this complex and rushed computer project.”.

In contrast, Skanska’s expert provided a less complex analysis, but the judge said as follows:

“He impressed me as someone who was objective, meticulous as to detail, and not hide bound by theory as when demonstrable fact collided with computer programme logic.”

The courts have been highly critical of experts providing testimony based on incomplete information. Expert witnesses must not accept facts / information presented to him without properly testing those facts. In Great Eastern Hotel Company Ltd v John Laing Construction Ltd [2005] EWHC 181, Laing’s expert produced an ‘impacted as planned’ delay analysis, impacting the planned programme with the alleged delays. However, the court rejected his evidence for various reasons, including as follows:

He has demonstrated himself to be lacking in thoroughness in his research and unreliable by reason of his uncritical acceptance of the favourable accounts put forward by” [the instructing party].

The judge also commented on the expert’s failure to revise his opinion in the light of facts subsequently becoming apparent, saying the expert “chose to ignore” documentary and photographic evidence made available to him “seeking to justify his erroneous analysis”.

Judge Wilcox also observed that the expert’s analysis “takes no account of the actual events which occurred on the Project and gives rise to an hypothetical answer”, saying that a “theory” relied on by the expert “collided with reality”.

Contemporaneous progress records are frequently relied upon when preparing a delay analysis. However, the judge in the Laing case said as follows:

It is evident in my judgment that Laing consistently underplayed mention of the true causes of critical delay and assert other reasons for delay that would not reflect upon them. They consistently misreported the delays actually occurring and manipulating the data in the programme update to obscure the accurate position”.

In the Scottish case of City Inn Limited v Shepherd Construction Limited [2007] CSOH 190, City Inn’s expert used an as-built critical path analysis, but it was found to be flawed because of the applied logic. In respect of this Lord Drummond Young said as follows:

“The major difficulty, it seems to me, is that in the type of programme used to carry out a critical path analysis any significant error in the information that is fed into the programme is liable to invalidate the entire analysis.”

When two opposing delay experts give evidence, the court obviously cannot agree with all of the conclusions of both experts, but there are also several instances of judges not accepting the conclusions of either expert. In PPG Industries (Singapore) Pte Ltd v Compact Metal Industries Ltd. [2013] SGCA 23 both parties’ expert witnesses “proffered diametrically opposed views as to the cause of the 273 days of delay”. The Singapore Court of Appeal concluded “both experts’ positions are unreasonable and cannot, by any measure of logic and common sense, be accepted in their entirety (if at all).”.

In the Australian case of In White Constructions Pty Ltd v PBS Holdings Pty Ltd [2019] NSWSC 1166, the opposing experts arrived at drastically conflicting conclusions, criticising each other’s exercise:

  • The claimant’s expert applied an ‘as-planned versus as-built windows’ analysis, with the defendant’s expert criticising it for assuming unjustifiable logic links, and being factually and analytically flawed.
  • The defendants’ expert applied a ‘collapsed as-built’ analysis, in respect of which the court observed: “This method requires the selection of “logic links” which link various components of the works to assume relationships of dependency to determine a critical path.’”. The claimant’s expert criticised this analysis for being, too simplistic, containing unsustainable and unjustified logic links, and hiding the defendant’s failures. It “assumes causation rather than identifies actual evidence of it”.

The judge decided that neither analysis was right. He also described their reports as impenetrable “to the unschooled”.

To help him decide the case, the judge appointed his own expert, whose evidence was “invaluable” and demonstrated “that the complexity that has been introduced is a distraction”.

In PBS Energo AS v Bester Generacion UK Ltd & Anor [2020] EWHC 223, the court concluded that one of the expert reports:

was based in part on documents whose genesis was obscure … and on the work of another expert who had neither signed the report or been put before the Court to be questioned. He was himself unable to answer questions on that other expert’s input”.

Experts should never forget the principles laid down in the Ikarian Reefer case, including that their duty to the court overrides any duty to their client. As most Expert appointments are by one of the parties, this is not always as straightforward as it may seem, especially given that the information the expert relies on has generally been provided by the appointing party. The problem of objectivity was highlighted in the recent case of Thomas Barnes & Sons plc (in administration) v Blackburn with Darwen Borough Council [2022] EWHC 2598, in which the judge said that one of the delay experts had:

formed a view as to the causes of critical delay from his instructions and his reading of the contemporaneous documents and the claimant’s witness statements, which he had then reverse-engineered into his fairly simplistic Gantt chart, subsequently modified when producing his report, without undertaking an open-ended analysis from first principles”.

Hence, when reviewing the information provided, the expert should apply some healthy scepticism, as the courts are not pleasant places when experts fail to meet the required standards.

CONCLUSIONS

Delay claims are now a major source of conflict in the construction industry, and also one of the most difficult to resolve. There are many different methods of delay analysis that can be used to demonstrate the causes of project delay.

For the uninitiated, there is a perception that specialised delay analysis software produces reliable conclusions about the causes of delay. This is perhaps even more the case when phrases like ‘forensic’ delay analysis are used. However, perceptions and reality do not always align! Regardless of which method of delay analysis is used, there are a plethora of ways to manipulate the outcome of a delay analysis exercise.

The authors of ‘Hudson’s Building and Engineering Contracts’ makes the point as follows:

there are obvious limitations on the effectiveness and utility of all of the methods. All rely to a greater or lesser extent on assumption, subjective assessment and theoretical projection. The answer provided by a particular method is thus only as good as the accuracy of the base information, the assumptions made and the reasonableness of the subjective assessments made by the analyst”.

When presented with a delay analysis exercise, regardless of their methodology, one should remember that things may not always be as they appear, and many lies sound like the truth, so one can have a hard time telling them apart.

I am not going as far as to say that parties to a construction dispute (or their appointed experts) tell lies, but they frequently present delay analysis exercises that provide incorrect conclusions regarding the actual causes of delay. This is not necessarily intentional; it could be the result of:

  • Weaknesses in the method chosen
  • A reluctance to invest a great deal of time and money into producing an analysis
  • Flawed assumptions / base line programmes
  • Flawed assessments / theoretical projections
  • Errors in logic links
  • Inaccurate progress records
  • A lack of as-built data

 

 

Previous Post
Take Notice! – A Brief Guide to NEC Contract Notices

Related Posts

Coniston Construction Associates Ltd - Specialist commercial and contractual services for the construction industry
keyboard_arrow_up