Construction disputes read differently from almost every other category of commercial arbitration. A shareholder dispute or a supply contract claim usually turns on a handful of documents and a legal question. A construction dispute turns on thousands of documents - site diaries, hindrance notices, revised programmes, minutes of joint measurement, variation orders, weather logs, labour deployment records - and on a question that lawyers are trained to argue but not to answer: what actually caused the delay, the cost overrun, or the defect, and by how much.
That is why construction arbitration is often described as “factually intensive.” The label is accurate, but it understates the point. In most construction references, the outcome is decided less by the tribunal’s view of the contract clause than by whose factual narrative the tribunal accepts on delay, disruption, and quantum. Experts are the mechanism through which that narrative gets built, tested, and ultimately adopted or rejected. Understanding how they are appointed, what weight their evidence carries, and where Indian practice still lags the international mainstream is essential for any attorney or in-house counsel handling an infrastructure or EPC dispute.
Why construction disputes need experts at all
Three categories of dispute drive the bulk of Indian construction arbitration, and each has its own expert discipline attached to it.
The first is delay and extension of time. Almost every infrastructure contract - highway EPC packages, hybrid annuity model concessions, metro and power projects - carries provisions for extension of time tied to employer-caused hindrances, force majeure, or changes in scope. Whether a contractor is entitled to an extension, and whether that extension also carries a right to prolongation costs, depends on establishing which activities were on the critical path and which delay events actually pushed completion. That is a scheduling exercise, not a legal one. It requires a delay analyst who can reconstruct the programme, identify the critical path at the relevant time, and isolate the effect of specific hindrance events from the effect of the contractor’s own inefficiency.
The second is disruption and loss of productivity. Even where a project finishes on time, a contractor may claim that employer-caused interference - repeated design changes, restricted access, out-of-sequence working - reduced output and inflated cost. Disruption claims are harder to prove than delay claims because there is rarely a single document that shows lost productivity; it has to be measured, usually by comparing actual output against an unimpacted baseline or against industry productivity data. This is squarely expert territory.
The third is quantum. Once liability for delay, disruption, or defective work is established, someone still has to price it - prolongation costs, idle plant and equipment, additional overheads, defect rectification costs, or the value of variations that were never formally instructed. Quantum experts, typically forensic accountants or quantity surveyors, translate factual findings into a number the tribunal can award.
None of this is naturally within a tribunal’s own competence, however experienced the arbitrators are as lawyers or even as retired judges. That is the gap experts exist to fill.
What Indian arbitration law actually says about experts
The starting point is Section 26 of the Arbitration and Conciliation Act, 1996, which gives an arbitral tribunal the power to appoint one or more experts to report on specific issues, and to require a party to give the expert relevant information or access to relevant documents, goods, or property for inspection. Once the expert delivers a report, either party can require the expert to appear at an oral hearing to be questioned, and can bring its own expert witnesses to address the same issues. The expert must also make available to the parties, on request, the material used to prepare the report.
Section 26 is permissive rather than mandatory. In practice, Indian tribunals rarely appoint their own experts in construction references; the far more common model - as in most party-driven arbitration internationally - is that each side instructs its own delay, disruption, or quantum expert, and the tribunal weighs the two competing reports against each other, sometimes alongside a tribunal-appointed expert on a narrow, genuinely technical sub-issue where the party experts cannot be reconciled at all.
The Indian Evidence Act, 1872 is not directly binding on arbitral tribunals - Section 19(1) of the Arbitration and Conciliation Act specifically frees tribunals from the Code of Civil Procedure and the Evidence Act - but its logic still shapes how Indian-qualified arbitrators and Indian-trained counsel think about expert evidence. Section 45 of the Evidence Act treats the opinion of a person “specially skilled” in a science, art, or technical field as a relevant fact precisely because the question is “beyond the range of common experience.” That is the same rationale tribunals apply informally when deciding how much weight to give a delay or quantum report: expert opinion is admitted and weighed not because a rule compels it, but because the tribunal has no independent means of resolving the technical question itself.
A useful comparator is Order XXVI Rule 10A of the Code of Civil Procedure, 1908, which allows a civil court to appoint a commissioner for scientific investigation where a question in the suit cannot conveniently be investigated by the court directly. Courts have used this provision for exactly the kind of technical fact-finding - surveys, measurements, valuation - that construction disputes routinely require. Arbitral tribunals are not bound by the CPC, but Section 26 of the Arbitration Act performs an analogous function: it is the tribunal’s own route to independent technical fact-finding when party experts leave a genuine gap.
The evidentiary reality: tribunals are the masters of expert evidence, and courts will not revisit that
The single most important structural fact about expert evidence in Indian construction arbitration is this: once a tribunal has weighed competing expert evidence and reached a factual conclusion, that conclusion is very difficult to disturb.
Indian courts have repeatedly held, including in Supreme Court decisions handed down as recently as 2025, that a court exercising jurisdiction under Sections 34 and 37 of the Arbitration Act does not sit as an appellate authority over the tribunal’s findings of fact. An arbitrator is treated as the master of the quantity and quality of evidence before them and is not bound by strict rules of evidence. Courts will not set aside an award merely because a different view of the expert evidence was possible; intervention is reserved for cases where the tribunal’s view is one no fair-minded person could have taken, or where the award discloses patent illegality or conflicts with the fundamental policy of Indian law. Courts do not re-examine expert witnesses or reweigh delay and quantum reports at the setting-aside stage.
This has a direct practical consequence that is easy to underweight when a dispute is still fresh: whatever the tribunal decides about which delay analysis is correct, whose productivity baseline is credible, or whose quantum model should be preferred, is close to final. There is no meaningful second chance to fix a weak expert case at the Section 34 or Section 37 stage. The entire contest over the facts happens once, in front of the tribunal, and the expert evidence presented there effectively is the case. Attorneys who treat expert selection and expert instruction as a secondary, administrative task - something to arrange after the legal pleadings are settled - are underestimating exactly the part of the case that Indian courts will not let them revisit later.
Delay analysis: the methodology fight that decides the case
Because delay claims are the most common source of factually intensive disputes on Indian infrastructure projects, the choice of delay analysis methodology deserves particular attention.
Internationally, the reference framework is the Society of Construction Law’s Delay and Disruption Protocol, now widely known simply as the SCL Protocol. It sets out a hierarchy of accepted methods - time impact analysis, which models the effect of a delay event on the programme at the time it occurred, using contemporaneous updates; windows analysis, which breaks the programme into periods and identifies the critical delay within each; collapsed as-built analysis, which removes delay events from the as-built programme to see what completion date would have resulted; and as-planned versus as-built comparison, the simplest but least rigorous method, useful mainly where better records do not exist. The Protocol’s central preference is for contemporaneous, time impact-based analysis carried out as close as possible to the delay event, rather than a reconstruction built years later for the purposes of the arbitration.
Indian construction arbitration has historically leaned on the less rigorous end of that spectrum. A meaningful share of Indian references still proceed on narrative delay accounts and retrospective as-planned versus as-built comparisons, reconstructed well after the event, rather than the programme-based, contemporaneous methods the SCL Protocol favours. That gap is now being actively discussed within the Indian construction law community, with recent industry conferences and professional collaborations specifically addressing how SCL-consistent methodology can be introduced into Indian practice. The direction of travel is toward greater methodological discipline, but attorneys instructing delay experts today should not assume a tribunal will supply that discipline on its own. The choice of method has to be made deliberately, defended on the record, and - critically - matched to whatever contemporaneous records the project actually generated, because a sophisticated methodology built on incomplete site records collapses just as easily as a weak one.
Concurrent evidence: an efficiency tool India has not yet adopted for arbitration
One procedural innovation gaining traction internationally is concurrent expert evidence, commonly called “hot-tubbing,” in which opposing experts are examined together rather than sequentially, allowing them to respond directly to each other’s positions in front of the tribunal rather than through counsel. The technique has particular relevance to construction disputes, where two delay or quantum experts often disagree not on the underlying facts but on methodology, and a side-by-side exchange can expose that disagreement far more efficiently than days of separate cross-examination.
India has moved on this in the court system - certain High Court patent and intellectual property rules now expressly provide for concurrent expert examination - but the Arbitration and Conciliation Act, 1996 has no equivalent provision, and its use in Indian-seated construction arbitration remains occasional rather than standard. Given the scale of documentary and expert material in a typical infrastructure reference, and the cost pressure on both employers and contractors in extended proceedings, concurrent evidence is one of the more obvious efficiency gains still underused in Indian practice. Tribunals have the procedural discretion to order it under Section 19 of the Act even without express statutory sanction; the obstacle is custom, not law.
The institutional picture: where construction arbitration is actually being seated
India’s institutional arbitration landscape has matured considerably, and each of the newer institutions has had to build its own approach to expert-heavy references. The Mumbai Centre for International Arbitration, the India International Arbitration Centre, the Indian Institute of Arbitration and Mediation’s Hyderabad centre, and the arbitration facilities emerging around GIFT City all now compete for exactly the kind of high-value infrastructure and EPC disputes that generate the heaviest expert evidence. For attorneys drafting arbitration clauses on new infrastructure contracts, or advising on a dispute already underway, the institutional choice increasingly matters for procedural questions like expert appointment timelines, document production for expert reports, and the tribunal’s power to manage a multi-expert record efficiently - not just for the traditional considerations of seat and enforceability.
A live disruption: the shrinking space for arbitration in NHAI contracts
No explainer on Indian construction dispute resolution written in mid-2026 can ignore a development that is actively reshaping where expert evidence in this sector will even be presented. A Ministry of Road Transport and Highways circular dated 12 January 2026 directs that arbitration will no longer be the default dispute resolution route for claims exceeding ₹10 crore under specified road-sector contracts - covering Build-Operate-Transfer, Hybrid Annuity Model, and EPC contracts awarded or administered by the Ministry and the National Highways Authority of India. Under the circular, high-value disputes are to be routed instead through a tiered administrative settlement and escalation process, with court litigation as the fallback if settlement fails; follow-up clarifications in late February 2026 extended operational guidance to field offices. Claims below the threshold may still go to arbitration, subject to fast-track constraints.
The immediate legal controversy - whether a government circular can unilaterally displace a dispute resolution mechanism written into signed, executed contracts without the contractor’s consent - is a live question the courts will need to settle, and one this publication will continue to track. But the development matters just as much for how expert evidence gets used going forward. An administrative settlement committee is not a tribunal, has no equivalent to Section 26, and will not run cross-examination of competing delay or quantum experts. That does not eliminate the need for expert input; if anything it raises the bar for contractors, because a claim now has to be documented and quantified to a standard that can persuade a departmental committee without the benefit of an adversarial hearing to test it. Independent quantity surveyors and delay analysts engaged early, to build a defensible record before a claim is even submitted, are likely to matter more under this framework, not less. Attorneys advising contractors on MoRTH and NHAI contracts should treat expert engagement as a pre-dispute discipline now, not a step that begins once arbitration is invoked - because for claims above the threshold, arbitration itself may no longer be where the fight happens.
What good expert practice looks like in an Indian construction reference
A few practical disciplines separate expert evidence that persuades a tribunal from expert evidence that gets discounted, regardless of which side commissions it.
Independence has to be real, not just declared. A quantum or delay report written to a predetermined conclusion, with the expert taking instructions on the outcome rather than the methodology, is usually visible to an experienced tribunal within the first round of cross-examination, and it damages the credibility of the underlying claim as much as the expert’s own standing. Tribunals that are the final word on fact - as Indian courts have made clear they are - notice this quickly and penalise it in the reasoning of the award.
Contemporaneous records matter more than sophisticated modelling. A time impact analysis built on programme updates that were actually issued and agreed during the works will usually beat a more elegant retrospective reconstruction built for the arbitration alone. Attorneys should be advising clients on record-keeping discipline - site diaries, hindrance registers, minutes of joint measurement, correspondence on instructed variations - long before a dispute exists, because that record is what the expert will eventually have to build on.
Methodology has to be disclosed and defensible, not just asserted. Given the continuing gap between Indian practice and the SCL Protocol’s preferred methods, an expert who explains why a particular delay analysis method was chosen for the available records - and is candid about that method’s limitations - carries more weight than one who presents a single number without engaging with the alternative approaches available.
Finally, experts should be briefed on the issues, not on the desired answer. Section 26 exists precisely because tribunals need technical opinions they can trust to be independent of party instruction. An expert report that reads as advocacy rather than analysis undermines the very reason expert evidence is admitted in the first place.
Takeaway
Construction arbitration in India is decided on facts before it is decided on law, and experts are the mechanism through which those facts get established. Because Indian courts will not reappraise a tribunal’s factual findings at the enforcement stage, the expert evidence presented during the arbitration is, in practical terms, the only chance to get the delay, disruption, and quantum case right. Attorneys and in-house teams handling infrastructure and EPC disputes should treat expert selection, contemporaneous record-keeping, and methodology discipline as front-loaded priorities - not administrative steps to arrange once the legal pleadings are settled. With arbitration itself now being displaced for high-value NHAI disputes above ₹10 crore, independent expert documentation matters earlier in the dispute lifecycle than Indian practice has traditionally assumed.
Position yourself where the legal industry looks for expertise - sign up on exlitem.com to get discovered by leading lawyers and high-value clients.
