Time Impact Analysis: Strengths, Limits, and Misuse

A contractor on a GCC infrastructure project receives a late instruction to redesign a pumping station six months into a two-year programme. Its delay expert runs a time impact analysis, inserts a delay event fragnet into the accepted baseline, and reports an entitlement of 47 days. The employer's expert, working from the same records, produces a windows analysis showing 22 days. Both call their method "prospective" and both cite the SCL Protocol. The tribunal is left to reconcile a 25-day gap that turns, on closer inspection, not on the facts but on how each analyst applied the technique. This is the recurring problem with time impact analysis (TIA): it is a powerful and widely endorsed method that is also, in practice, among the most frequently misapplied.

WHAT TIME IMPACT ANALYSIS ACTUALLY IS

Time impact analysis is a modelled, cause-and-effect technique. The analyst takes a logic-linked critical path programme, introduces a delay event as a sub-network (commonly called a fragnet) at the point in time the event arose, and recalculates the schedule to measure the impact on the contract completion date. The output is the predicted delay attributable to that event, assessed against the programme as it stood immediately before the event occurred.

The Society of Construction Law Delay and Disruption Protocol identifies TIA as a prospective method: it looks forward from the moment of the event and asks what delay the event was likely to cause, rather than looking back to measure what delay in fact occurred. That prospective character is central to both its strengths and its limitations. The Protocol's core position on entitlement is that, for extension of time purposes, the analysis should reflect the likely effect of the delay at the time it arose. TIA aligns naturally with that principle.

WHY IT IS FAVOURED

The appeal of TIA is real and should not be dismissed. First, it maps directly onto how most standard forms operate. Under FIDIC and NEC, extension of time is assessed by reference to the effect of an event on completion, and both contemplate assessment close to the time the event arises. NEC's compensation event mechanism, with its quotations based on a forecast of impact, is essentially a contractual form of prospective analysis. TIA gives commercial and contractual effect to those provisions.

Second, TIA isolates individual events. Where several claims are in play, the technique allows each to be modelled discretely, which assists in attributing responsibility and in dealing with events sequentially as they arise. This is particularly valuable in contemporaneous assessment, when parties are administering entitlement during the works rather than reconstructing it years later.

Third, because it is grounded in critical path logic, a well-constructed TIA can demonstrate not merely that an event occurred but that it affected the completion date through a genuine chain of causation. That is more persuasive than assertions of delay unsupported by network analysis.

THE LIMITS THAT ARE OFTEN OVERLOOKED

TIA's prospective nature is also its principal weakness. Because it measures likely impact rather than actual impact, its results can diverge from what actually happened on site. A fragnet may predict 30 days of delay, but if the contractor mitigated, or if a subsequent event overtook the works, the actual delay may be very different. When abundant as-built records exist, a tribunal may reasonably ask why a forecast should be preferred over the record of events that in fact unfolded. The SCL Protocol acknowledges this tension: where good records permit a retrospective assessment, the actual position should not be ignored simply because a prospective method was chosen.

The method is also acutely sensitive to the quality of the baseline programme. TIA is only as reliable as the schedule into which the fragnet is inserted. If the baseline is logic-poor, contains open ends, relies on constraints rather than genuine dependencies, or does not reflect the actual intended sequence of works, the impacted result is unreliable regardless of how carefully the fragnet is built. Many disputes over TIA are, in truth, disputes over the integrity of the underlying programme.

A third limitation concerns the choice of programme to impact. In a rigorous TIA, each event is inserted into a schedule updated to reflect progress immediately before that event. Where multiple events are analysed, this requires a series of updated programmes, each reflecting actual progress to date. This is demanding and record-intensive. It is precisely where practitioners take shortcuts.

WHERE MISUSE CREEPS IN

The most common misuse is impacting the original baseline for every event, ignoring actual progress. This treats the project as though it proceeded exactly as planned until each event struck, which is almost never the case. It also allows an analyst to claim delay against a completion date the works were never going to achieve. If the contractor was already in culpable delay, or was ahead of programme, impacting an unadjusted baseline produces a figure divorced from reality.

A second form of misuse lies in fragnet construction. The delay sub-network is where judgement, and manipulation, live. The number of activities, their durations, and above all how they are logically tied into the existing network determine the result. A fragnet linked to a near-critical path can be made critical, or kept off the critical path, by adjusting a single link. Analysts should be able to justify every logic tie in a fragnet by reference to the facts and the records, not to the desired outcome.

Third, TIA is frequently misused to disguise concurrency. Because it isolates a single event, TIA can present an employer-risk event as the sole cause of delay while a concurrent contractor-culpable event is quietly absent from the model. Isolation is a strength only if the analyst engages honestly with what else was happening in the same window. Concurrency must be addressed on the facts, not defined out of existence by modelling choices.

Finally, there is the misuse of labelling. Analysts sometimes describe a heavily retrospective, results-oriented exercise as a "time impact analysis" to borrow the method's respectability, when what they have actually done bears little resemblance to a disciplined prospective assessment. The label matters less than the substance; tribunals increasingly look past the name to the mechanics.

USING TIA RESPONSIBLY

TIA is at its most defensible in near-real-time assessment, during the works, when the baseline is validated, progress is properly recorded, and events are assessed as they arise. It is less suited to complex, long-tail retrospective disputes with rich as-built records, where windows analysis or an as-planned versus as-built comparison may better reflect what actually occurred. The AACE International recommended practice on forensic schedule analysis is helpful here: it catalogues the methods, distinguishes observational from modelled and prospective from retrospective techniques, and stresses that method selection should follow the records available and the question to be answered.

The disciplined practitioner should therefore validate the baseline before doing anything else, update it to reflect actual progress before each event, justify every fragnet link on the evidence, address concurrency openly, and sense-check the modelled result against the as-built record. Where forecast and actuality diverge sharply, that divergence should be explained, not concealed. Method should serve the facts. TIA is a good servant and a poor master; used with discipline it is among the most persuasive tools available, and used carelessly it is among the easiest to discredit.

REFERENCES

1. Society of Construction Law, Delay and Disruption Protocol, 2nd edition (February 2017).
2. AACE International, Recommended Practice No. 29R-03, Forensic Schedule Analysis.
3. Keith Pickavance, Delay and Disruption in Construction Contracts, 4th edition.
4. FIDIC Conditions of Contract (Red, Yellow and Silver Books), 1999 and 2017 editions.
5. NEC4 Engineering and Construction Contract.
6. Keating on Construction Contracts (Sweet & Maxwell), current edition.