Is computer-based critical-path analysis essential when determining an extension of time?
No single software model or delay-analysis method is legally mandatory in every case. The decision-maker must nevertheless establish, by a logical and evidenced process, whether relevant events delayed completion and by how much.
Computer modelling is not an end in itself; the assessment must reliably identify critical delay to completion from the contract, programmes, progress records and actual events
A complex dispute will often require critical-path analysis, and software may be the clearest means of testing logic, float and changing criticality. A model built from inaccurate data or unsupported assumptions carries little weight. Thomas Barnes confirms that a tribunal should not decide the case by asking only whether an expert selected or followed a named method. It must assess whether the method suits the facts, whether departures are explained and whether the conclusions withstand the contemporaneous evidence and practical causation analysis.
| Issue | Position |
|---|---|
| Computer software | Useful but not universally mandatory |
| Complex interacting delay | Structured critical-path analysis will often be necessary |
| Named methodology | Does not decide reliability by itself |
| Contemporaneous records | Must test and support the model |
| Ultimate question | Which relevant event delayed completion, when and for how long? |
The contractual question comes first
The decision-maker must apply the extension-of-time clause: identify the qualifying event, determine whether it affected completion and assess the resulting period. The contract may prescribe information, timing or a prospective assessment, but it rarely makes a particular software package the source of entitlement. The analysis serves the contract rather than replacing it.
Criticality is a factual relationship
An event delays completion only if it affects work controlling the completion date, directly or through its successors. The critical path may change as work is resequenced, float is consumed, mitigation occurs or different activities become controlling. A single tender programme may not describe the critical path throughout the project.
Logic links, durations, calendars, constraints, progress status and the factual sequence must be supported and tested.
Balfour Beatty: assess the completion effect
The administrator should assess whether a relevant event delayed progress and, if so, determine the amount by which the completion date should be postponed. The task concerns completion causation, not merely the duration of the event itself.
John Barker: avoid impressionistic assessment
An architect was criticised for not analysing methodically how relevant events affected the planned programme, and for making an impressionistic assessment instead. A fair conclusion requires a reasoned path from evidence to the time awarded.
Royal Brompton: examine what was critical then
An assessment of ultimate completion should consider which operations were critical to progress when the event occurred. That enquiry is commonly assisted by critical-path analysis, but the required proposition is factual criticality rather than use of a particular computer tool.
City Inn: modelling is not invariably essential
The Scottish court rejected the proposition that a meaningful opinion could only be given through an as-built critical-path analysis. It warned that a material input error could undermine the programme model and endorsed a fact-sensitive, common-sense assessment.
Thomas Barnes: method is subordinate to reliability
The experts selected different delay methods and each was criticised for its implementation. The TCC held that excessive weight should not be placed on whether a named method was followed exactly. The court examined the suitability of the method, explanations for departure, contemporaneous documents, actual events and whether the resulting conclusions were practically sound. A detailed records-based analysis was preferred to a simplified chart that had not tested all potential critical causes.
Select a method suited to the question
| Approach | Use and limitation |
|---|---|
| Prospective time-impact analysis | Tests the forecast effect of an event using an accepted contemporaneous programme. |
| Time-slice or windows analysis | Examines changing criticality across defined progress periods. |
| As-planned versus as-built | Compares intended and actual timing, but requires careful causation analysis. |
| Longest-path review | Identifies the controlling as-built sequence, subject to logic and event proof. |
| Collapsed as-built analysis | Removes events from an as-built model to test a counterfactual, with sensitivity to model assumptions. |
| Factual narrative analysis | May be adequate for simpler events if the critical completion effect is demonstrated from records. |
Minimum reliability checks
| Check | Question |
|---|---|
| Baseline validity | Was the programme accepted, achievable, logically linked and consistent with scope? |
| Progress accuracy | Do updates reflect actual starts, finishes, remaining durations and resequencing? |
| Critical path | Is the controlling sequence identified for each relevant period? |
| Event insertion | Are dates, responsibility and programme effects supported by records? |
| Concurrency and float | Are parallel causes and available float addressed? |
| Sensitivity | Would reasonable changes to disputed inputs alter the conclusion? |
Administrator and expert roles
A contract administrator may have to decide within a short contractual period and on incomplete information. The decision must still be fair, reasoned and based on the material available, with requests made for missing particulars. A later expert analysis may use more records and different tools, but should not be confused with the information reasonably available at the earlier assessment date unless the contract permits a later retrospective review.
Practical controls
Maintain a logic-linked baseline and contemporaneous updates.
Record actual progress and delay events before reconstructing the model.
Choose the method to answer the contractual and factual question.
Explain all assumptions, constraints and departures from the stated method.
Test alternative causes, concurrency and changes to the critical path.
Reconcile model conclusions with diaries, photographs, instructions and resource records.
Present a reasoned determination capable of being followed without relying on software output alone.
Authorities
| Authority | Citation | What it decides |
|---|---|---|
| Balfour Beatty Building Ltd v Chestermount Properties Ltd | (1993) 62 BLR 1 | The administrator must assess the amount by which relevant events delayed completion. |
| John Barker Construction Ltd v London Portman Hotel Ltd | (1996) 83 BLR 31 | An impressionistic extension assessment was criticised for lack of logical programme analysis. |
| Royal Brompton Hospital NHS Trust v Hammond | (2001) 76 Con LR 148 | The assessment should identify operations critical to completion when the event occurred. |
| City Inn Ltd v Shepherd Construction Ltd | [2010] CSIH 68 | As-built critical-path analysis was not treated as the only means of forming a meaningful opinion. |
| Thomas Barnes and Sons plc v Blackburn with Darwen Borough Council | [2022] EWHC 2598 (TCC) | Method labels were subordinate to suitability, explanation, records and sound factual conclusions. |