Which programs to be used for delay Analysis - baseline or updated

Hi Toby

Agreed for forensic analysis - but how it is demonstrated is always open to challenge.

For construction in progress nobody knows what the completion date will be.

Best regards

Mike Testro

Charlie

Actual delay is not a common clause, but can be found in ICE 7, which, depending upon how it is interpreted, may (and I stress ‘may’ because the phraseology is far from easy to construe) restrict analysis for the purposes of entitlement to an extension of time to a calculation based upon actual delay to completion alone.

Toby 

Hi Aqil

You seem to be using the American definition of TIA and IAP.

Every where else the definition is different.

Best regards

Mike Testro

1. Usage of TIA is no doubt more practical and therefore more accepted these days, also this is recognized technique by SCL.

2. TIA shows realistic scenerios to calculate EOT's for the encountered delays.

3. I believe IAP cannot be used for EOT's calculations as this is based solely on Baseline logic which may contain many flaws.

Thogh this BL may be approved in advance but when it comes to EOT's actual status based on updated program and logic/durations/lags/leads etc of remaining activities should be agreed between two parties before granting any EOT.

4. Based on the above i would recoomend everybody to use TIA , specially for big projects to be in a better position to defend their entitlement.

regards

Aqil

Toby,

A good summary. I would add the "Prospective results can be inaccurate" limitation to impacted as planned.

Also is i am not aware of any contract here that requires an actual delay to completion. Is that a common clause in the UK contracts?

The method is rarely specified here in Australian contracts. It is more implied in the wording of the contract. Ie it states something like "contemporaneous assessment". This precludes IAP most of the time.

Toby,

A good summary. I would add the "Prospective results can be inaccurate" limitation to impacted as planned.

Also is i am not aware of any contract here that requires an actual delay to completion. Is that a common clause in the UK contracts?

The method is rarely specified here in Australian contracts. It is more implied in the wording of the contract. Ie it states something like "contemporaneous assessment". This precludes IAP most of the time.

Dear All

Some points to note regarding delay analysis:

• Likely or Actual delay to completion
• Method Specified

• EOT or compensation or both

• –Availability and Integrity of an As-Planned Programme , progress and/or As-Built Data;

• Small disputes do not justify expensive and time consuming analyses

For Impacted as Planned:

Advantages

• Fairly quick and easy to carry out
• Easily understood
• It can be used where as-built information is limited/ does not exist
• Can be suitable method of proof if:
  • Project is of simple character
  • The planned programme was realistic and achievable
  • The critical path remains largely unchanged except for the effect of events

Limitations

• Ignores the as built history of the works
• Cannot deal with concurrency
• Ignores the effect of any change in sequence or acceleration
• Takes no account of:
  • Progress
  • Resources
  • Changing logic

For Time Impact Analysis

Advantages

• Cause and Effect
• Takes account of Progress/Resource/Logic
• Is the most accurate because it uses all planned, progress and as-built information
• Takes account of inadequate progress
• Takes account of changes in methodology and re-sequencing
• Can be used to resolve concurrent delay
• Takes account of  acceleration
• Can deal with multiple Key Dates & Milestones

• Requires high quality information
• Can be time consuming and expensive
• Produces a high volume of output
• Complicated (and therefore slow)
• Prospective results can be inaccurate
• Can be difficult to communicate (Skanska v Egger[2004])

Andrew,

I agree with your comments. Even when the project is over and then the dispute raises its head months or even years later, the assessment of the delay claims should be carried out prospectively (not restrospectively).

You would put yourself in the position of the contractor at the the time the delay occurred. At what point in time that is may be the subject of further dispute.

In the case of a variation, i would say when it was formally instructed. As you know its never this clean cut as often variations are progressed based on series of verbal or informal directions. 

Regardless, the estimated impact on program of a variation is not an actual impact but a theoretical one (but then costs are theoretical also- unless it is on a schedule of rates or fully reimbursable). I.e. in an ideal world a variation is issued by the client, the cost is estimated (+$200k) and impact to program (+5d). The client is advised and approves the variation.

If the program impact ended up being only +4d and cost +160k you would not expect the client to ask for time and money back would you? 

Conversely, all else being equal, if the time overran (+1d, i.e. 6 d in total) and so did the costs  (+40k, i.e. $240k in total) for carrying out the variation , then the contractor wears this overrun as it has not managed the execution of  variation within its agreed estimate (cost and time).

9 times out 10 from my experience the contractor quotes the variation cost and is silent on the time impact. Unless you have a competent Superintendent (or Contracts Administrator) who specifically asks for an estimate of the time impact ALSO to understand program risk, the program impact aspect is overlooked. A big mistake. 

Then there is a real chance that after the work is carried out , or part way through it, the contractor hits the CA with an EOT for the variation and the fun begins. It never ceases to amaze me how often this repeats itself.

NB: If you are able to estimate the cost of a variation looking forward , you should be able to estimate the impact on program (otherwise how do your price a variation when you haven't properly assessed the program impact?).

not enough time, so i should have said nothing. please accept my apologies.

Hi C Andre

Please elaborate on your brusque comment - or I will have to delete it as being offensive.

Mike Testro

Interesting topic although not an expert in the field I think we need to consider how the programme should be managed throughout the course of a project.

1. Prepare the original programme, Contract, Master whatever the form of contract calls it. (this should be done in the form prescribed by the contract)
2. Assuming you are using computer aided method save a copy as baseline (target)
3. Start the contract
4. At regular intervals update the current programme with progress and reschedule the programme
5. When the forecast completion date is scheduled to occur later than the original completion date a "critical delay" has occured and under most forms of contract it should be reported to the client. If the delay is due a factor outside the contractors control (compensation event) covered by the contract then an extension of time can be applied for. If it is a delay caused by the contractor then the contractor will need to re programme to recover the delay.

I am sure Mike and Charlie will correct me but if this proceedure is followed the need for forensic analysis is removed, however that is not how it works in real life is it?

My understanding that forensic analysis should take account of progress at the time of the delay to be fair to both parties in the contract.

Charlie > Mike

i'd go with Charlie's method.

Mike,

Thanks. I do have a copy of Pickavance and have read parts of it.. not light bedtime reading as you know. The parts i have read that deal with methods is pretty explicit in giving the TIA the thumbs up. I'lll go back a read the  early cases more throughly.

Perhaps in the past decade the tide is shifting to TIA? I have worked on just over 20 claims (that have gone to court/arbitration) over the last 12 years. All of these disputes have all been won using the TIA method (other than a few claims with disruption aspects). In about a third of the cases, we were required to discredit (with 100% success rate) the IAP method used by the opposition. Maybe things here on Australia are different.

Where i have done IAP analysis it is for contractors seeking a commercial resolution to the dispute rather than a more expensive and onerous litigation process. There's no doubt its simplicity is attractive.

Cheers,

Charlie

Hi Charlie

If you read Kieth Pickavance's book on delay analysis you will find that his first cases were all based on the Impacted as Planned method of delay analysis - even in forensic situations.

Ten to fifteen years ago it was the only method used in the Uk until methods got more sophisticated.

All my adjudications were won on IAP methods up to 5 years ago when the SCL protocol became better understood by the tribunals.

Best regards

Mike Testro

Mike,

I sent that last one off half baked. Anyhow no big deal. Thanks ans may i add same to you!

I am genuinely interested to know of any cases or case studies for the IAP method being sucessful. Can you refer me to any that you are aware of? Whether in the Uk or elsewhere it doesnt matter.

Thanks

Charlie 

Hi Charlie

You just carry on doing it your own way - I know its wrong but you disagree.

I hope it works for you in your next arbitration.

Best regards

Mike Testro

Hi Mike,

Of course i agree  in the Impacted as Planned there are no progressed tasks simply planned tasks- of thats what you meant.

Other than thats looks like we have very differing opinions on this!!

Re your response to my questions: i believe it is irrelevant if the the contractor is falling behind the baseline program, this does not invalidate the current program as a tool for the starting point of an EOT assessment.

Consider the following scenario: the baseline program shows PC on the agreed contract date of 300. Then 6 months later the program forecasts a  completion on day 330 (due to contractors own progress. Then at 6.5 months EOT Delay Event arises. It is found to have pushed completion out to day 350 (ie a nett 20 day delay). Contractor gets an EOT from 300 to 320, obviously not 300 to 350. 

The fact that the current program is not a promise or an acceped baseline is likewise irreleavant.  If delays (client non claimable and contractor's own delays)have been experienced since the baseline was developed then it should rightly have been recorded in the program as part of the updateing process.

What is important is that the current program is a reaosnable representation of the contractors position just prior to the delay and a reasonable representation of how it intendeds to execute the works. The CA would have to have a very good reason for rejecting the current program as the basis for an EOT assessment. Even if it aspect of it are found to be deficient then, once the programe is corrected to address these deficiencies then the program would be good enough, ie no need to revert to a 6-9 month old program.

Th

Hi Mike,

Thanks for the reply, I really appreciate it.

For sure there will be disputes because of the presence of subcontractors;

In this case, who will be responsible for reconstructing the baseline schedule?

The one who makes the claim, or the one defending it?

So if it is in the impression that the reconstructed baseline might be or will be rigged, then maybe both parties (the claimant, and defendant) have the right to do their own schedule?

cheers

Hi Anoon

If you have not got a valid baseline programme then it may be possible to prepare one retrospectively - in fact the SCL recommends this approach.

In my opinion it is a very bold analyst who sets out to reconstruct a baseline programme using hindsight because he/she will be challenged that the new baseline has been rigged.

If I was asked to undertake such a task I would decline the commission.

In reality if neither the contractor or the employer is that bothered about progress or delays then there will be no dispute.

Best regards

Mike Testro

Hi Charlie

Short reply:

1 why would you revert to a baseline program potentially 6-9 months old as a basis for analysing delays when you have a current (further development) of it at your disposal?

Because it will be the last true representation of the contractor's promise to complete on time. If there have been further developments and adjustments that have crystalised into a new approved programme then that will do - a progressed chart cannot be impacted on because it represents the difference between what was promised and what was achieved.

2 If the current program does not provide a better tool to manage the project, including assessing the impact of various scenarios (including own and client delays) than the older baseline program- why bother to update the baseline at all? 

The updated programme shows what is actually happening INCLUDING the effect of any delay - you will be impacting on an impacted programme. The progressed (As Built) programme is used to compare the theoretical impact of the delay with what actually happened on site.

In a true Impacted as Planned there are no as built records because the work has not yet started.

Best regards

Mike Testro

Charlie & Mike,

just some questions (if I may?):

1. What if there are no available programs or schedule at all?

I mean the Contractor just do the job using their own schedule which was never approved by the Client;

And the Client just let the Contractor do the job, without bothering whether the Contractor is ahead or behind schedule;

please consider the following:

1. given a somehow long contract duration - presumably to contain the long-lead items to be supplied by the Client;

2. that's why the Client does'nt bother about contractor's delays (if any, at the current time);

3. because the Client is anticipating their very own delay as well;

then what program or schedule to use?

Hi Chris,

Australian Standard Contracts (AS2124 & AS4000) require that contractors submit an EOT as soon it becomes aware of a delay (not a notification of likely delay), not just at the cessation. In some cases, they are required to continually notify the Contract Administrator (CA) in writing on regular basis (eg each week) if the delay is ongoing. This can be a pain in the backside.

In cases where the end of the delay is hard to estimate (eg a hold on a section of the work s by the client) then you would of course need to wait to the end of the delay to work out the impact and submit a final EOT.

However in the case of a variation, say the Contractor (C)  was building a 30 level building and early on in the project a further 5 floors were added to the scope, then this could be done forward looking without needing to wait for the works associated with the delay event to actually end.  In this case there is no need to assess the actual impact on the project that eventuated (although CA's sometime like wait and see what happens as it means holding off longer to grant any time).  In AS2124 it is the CAs responsibility to assess the EOT as soon as possible (to the delay event ) so the contractor can be appropriately compensated and re-plan/re-prioritise its work accordingly. An example of replanning might be not pushing feverishly on works that were previously critical and now are non-critical as a result of the client delay and corresponding EOT.

The delay event occurrence in this case is not when the works are physically carried out from Levels 31-35 but when CA was issued the Variation to the C requesting it to proceed with the works.

Apart from the cessation of an ongoing delay (such as a hold), there is nothing I have seen to suggest that the CA should wait and see what happens (i.e. award the time retrospectively based on the outcome of actual events) in any construction contracts in Asutralia .

So yes the method is imperfect and hypothetical but it is the best hypothetical method we have so far.

Regards,

Charlie

Hi Mike,

The process for TIA could also be done on pen and paper without a computer, though I would not like to try it. The TIA is a simple process also, albeit in a project with dozens of delays cannot easily be assessed on a single impacted program. It is considered the most robust method according to the UK SCL and other experts (e.g. Pickavance) to withstand scrutiny in courts.

 The starting point of a delay should be the contractor’s position at the time the delay occurred. For reasons explained earlier, not some redundant position it was in months earlier.

The likely effect  on completion date cannot be calculated as accurately using the (impacted as planned (IAP- per Picakavance’s  description and mine in earlier posts here)  as it fails to take into account contemporaneous events, including the critical path as close the to the time the Delay event occurred. Additionally the delay events are not usually inserted in the program at the time they occurred (i.e. on the date they occurred) but slotted somewhere in the sequence.

The IAP seems to be universally recognised (UK, USA and AUS) as the least robust analysis under cross examination and far less logical then TIA (although the IAP is most simplest &  cheapest to prepare hence why contractors prefer this method).

I have prepared IAP for clients (even though I have advised against it) for this very reason in the past but now am aware of its deficiencies would avoid using it. Where the opposing side  has used this method (either during the course of the project or forensically), we have been able to discredit it successfully.

 A progressed program (save for objectionable deficiencies) is the best program to measure delays for all of the above reasons.

Some questions for you:

1 why would you revert to a baseline program potentially 6-9 months old as a basis for analysing delays when you have a current (further development) of it at your disposal?

2 If the current program does not provide a better tool to manage the project, including assessing the impact of various scenarios (including own and client delays) than the older baseline program- why bother to update the baseline at all?

Regards,

Charlie

Hi Christopher

It sounds as though you have been using the AACE as bed time reading.

There is a whole industry - mostly emanating from USA - which endeavours to complicate what is really a very simple process.

Delay analysis is not rocket science - it is the application of experience and common sense so as to try to explain what caused the delay and what the effect was.

It does not need a barchart or baseline programme to do this succesfully - before computers arrived it was a simple graph paper and pencil operation - and I still sometimes use a pure narrative presentation because some tribunals prefer it.

My basic principle is to keep it so simple that even a Judge can understand it.

The starting point of any analysis must be what the Contractor promised to achieve prior to the delay event and the agreed method of achieving it.

The likely effect of the delay event on the completion date can then be calculated.

A progressed intermediate programme has no part in the process.

Best regards

Mike Testro

Hi Joseph

It is very simple:

For work in progress use the Time Impact Analysis  and for forensic / retrospective analysis use Impacted as Planned (UK Definitions)

The As Planned v As Built method works well in Sub-Contract situations but will not show direct cause and effect in main contract situations and should therefore be avoided..

Windows analysis has now been overtaken as a valid method for delay analysis.

Best regards

Mike Testro

Dear All,

This is a good conversation and knowledge input for every one, I my self is having the same question  but on a different aspect

the Question is if which of the Methods is the most appropriate to use in Delay Annalysis.

Currently we are having a project that was only at engineering stage but already having delay due to clients late information on the design criterias to be used in completing the design.

Now at this stage of the project I believe that the most appropriate method to use is IAP (Impacted as Planned) with the use of original baseline schedule and adding the activities that cause the delay and link it to the activities that it will affect and status without changing the Baseline Data Date.

But if the delay happened in the middle of progressing schedule where there are already involvement of Owner and Contractor delay i think this should be presented using Planned Vs As Built method, but to be supported by a proper statement and arithmetical presentation if which one is delayed due to owner and which one is delayed due to contractor., or this is better known as Window/s Annalysis Method.

this is just my idea and i actually need also input if this is a correct approach or no.

and if you have any format or template for extension of time i would like to ask if you can send it to me thru my email Use Private Message Service

Thank you and Best Regards,

Joseph Cargason

Hi Charlie

First to definitions in UK and USA.

The SCL protocol's method that deals with delays during the progress of the work is called Impacted as Planned. In the AACE the same method is called Time Impact Analysis - total confusion.

It remains my opinion that you should not use a progressed programme as the basis to impact delays while work is in progress.

If the logic of the original Contract/Baseline programme has been changed then a new Baseline should have been submitted for approval which would then become the current Baseline Programme and delay events should be impacted on that version.

Best regards

Mike Testro

Thanks Mike,

Can you please refer me to a comparison of the US and UK TIA methods. I was not aware there were major differences.

In a forensic situation in Australia the claim is best viewed from the position of the contractor at the time the delay occurred (i.e. a contemporaneous assessment). This is not retrospectively reconstructing the actual impact of the delay but making an estimate of the impact of the delay looking forward. I understand form the UK SCL Protocol that this is the preferred method there and, from what i have read, in the US also.

Using the original baseline/contract program would only be appropriate if you are still following the exact methods described in the baseline program and you are keeping on track to program (including non critical works). This is unlikely to be the case on most projects for the following reasons:

1.    Methods change (plant/resources/sequence)

2.    Non critical works are not progressed by early dates (it is within the contractors rights to delay non critical works if it wants to) and therefore float changes, including contractors own delays in design/procurement and construction.

3.    Client delays which are non critical yet have an impact on float paths need to be taken into account.

In short the courts are interested in understanding the contractor’s position just before the delay (or as close as possible as records will allow you to reconstruct) occurred. This would not be best represent by an outdated baseline/contract  program which would be potentially months old, which does not take into account the changed situation (due to reasons i mentioned above), but instead would be best represent by the current updated program.

Re distortion, the updated programmes would not give a distorted result as we are only trying to establish the nett impact on the end date. So any contractor delays up to the client delay would be included in both the before and after program (as it should be - and hence not unfairly advantage the contractor in its claim). Likewise any acceleration employed by the contractor considered and incorporated into the (before) program up to the time of the delay will be taken into account in the after scenario and provide a fair assessment.

Hi Charlie

First you have to tell us which definition you are using for TIA - American or UK because they are quite different.

In my opinion you should not be using updated programmes to impact an entitelement to an EoT because the updated programmes may include Contractors delays or increased performance that would then be wrapped up in the effect and give a distorted result.

You must use the original contract / baseline programme.

If it is a forensic situation and the event has already caused delays then as I said earlier in this topic you must use the updated version to compare with the impact on the original programme.

Best regards

Mike Testro

Hi Alaa,

It depends on when the delay event occurred and also the quality and level of record keeping for as built dates. 

Generally the program which was current prior to the delay event occuring is the correct one to commence the analysis with. In most cases this will be the updated construction program (Time Impact Analysis method). 

TIA example:

1. The project is 30 months long
2. you are providing program updates to the client monthly 
3. At 20.5 month into the project a variation is raised (formally instructed) by the client causing a delay.
4. You would use the program statused at the end of month 20 as the baseline (i.e. the before scenario)- call it program M20
5. Modify M20 for the estimated effect of the variation. Do not include any other changes as it will muddy the analysi. What you are trying to do is isolate the impact of the delay event on program.Include any design and approvals, material procurement etc. that is required due the variation.
6. The resulting impacted program is called M20A
7. Compare the result of M20 and M20A to detemrine the impact on PC , specific milestones or any Separable Portion dates.
8. NB the above is based on the premise that the M20 "before" program has no major logical/duration/resource flaws and is a reasonable representation of how the contractor intends to execute the program. If there are no major logical/duration/resource flaws then the program is a good basis (baseline) on which to commence the  analysis. If there are flaws the client or its rep can reject ther program as basis for the analysis subject to corrections which are reasonable.
9. I have worked on claims where we atttacked the credibility of the before program. In once case the updated program did not allow sufficent time per the specification for curing and hydrostatic testing of concrete a pool shell. Once the program was corrected for this the critical path shifted to the pool construction and impacted area subject to of contractor's claim was no longer critical .
10. The above method is far superior to the impacted as planned (see below) which is less robust under analysis in cross examination in courts.
11. The TIA (time impact analysis) method requires generally 2 programs for each delay assessment (a before and after program). As such on a project with 10 EOT claims, 20 programs can be alot of work and seen as too complicated for many contractors. In my experience they prefer the Impacted as planned (IAP) as it  quick and cheap to produce. It is a single program with all delays plugged into it.  However the IAP method is severely flawed. Read more at UK SCL and AACE (US body). Pickavance also.

The initial contract "baseline" would only be used as a baseline program in the following circumstances (impacted as planned method):

1- it was contractually specified that the impact as planned method would be used to assess delays. I have never seen it sepcified personally.

2-the were virtually no as built records kepts during the course of the project with which to determine the status just prior to the delay event occuring . This scenario is hard to imgaine nowadays on large projects with the plethora of record keeping requirements.

3- The delay event occured in betweent the baseline being developed and prior to any subsequent updates. In my opinion this is really the only legitimate time and reason to use the IAP method.

4- This method (impacted as planned) is preferred by builders as it allows all delays (over the life of the project) to be insterted into a single impacted as planned program to generate a result. This method is largely flawed as the project critical path varies throughout the course of a project and this method also fails to take into account the status of works just prior to the delay including the builders entitlement to use up float on non critical areas.

Hope this helps.

Charlie

Alaa,

In my opinion, both methods are correct, depending on the purpose.

I think what your friend was trying to tell you is to analyse the delay by  including  a new activity (cause) to the baseline programme,  link iit correctlly to the proper successor activity, then schedule (F9) based on  baseline data date to analyse the delay (effect).

What you have in your mind is also correct, by comparing an updated programme and compare it to the baseline plan. In this case your analysis is progress, which activities have not started, or did not achieve the planned progress, or actual start dates did not start as planned or finished as planned, those sorts..Then you determine what is causing those delays, either it is lack of materials, manpower, submittals, NOCs, approvals, etc.

Hi Alaa

You need to start with the original programme to impact the events and compare the results with the actual progress or As Built data. So in effect you have to use both programmes.

There are other sources for as built data and all the information has to be combined to get an accurate picture.

Best regards

Mike Testro