Concurrent vs Parallel Delay in FIDIC Projects: Technical Distinctions, Legal Risks, and Tribunal Perspectives

Concurrent vs Parallel Delay in FIDIC Projects: Technical Distinctions, Legal Risks, and Tribunal Perspectives

1.0 Introduction: Why This Distinction Matters

In FIDIC-based construction projects, especially those operating under international contract standards in civil law jurisdictions like the UAE, few concepts are more frequently misused, or strategically distorted than concurrent delay and parallel delay.

The misunderstanding or mislabelling of these terms often leads to:

• Unjustified denial of Extension of Time (EOT) claims,

• Improper enforcement of liquidated damages under Clause 8.7,

• Misallocated entitlement to prolongation costs, and

• Costly disputes that escalate unnecessarily to adjudication or arbitration.

This is not a question of semantics, the classification of delay events has a direct legal and financial consequence. When a contractor submits an EOT for Employer-caused delay, the Engineer or Employer often replies, “There was also a Contractor delay in that period; therefore, no EOT.”

This response implies concurrency — but was it really concurrency?

In practice, many overlapping delays:

• Do not both affect the critical path,

• Are not independently causative,

• Or stem from the same party.

As this article will demonstrate, the distinction between concurrent and parallel delays is not just technical, it is determinative. That distinction decides whether:

• An EOT is granted in full or reduced,

• LDs are enforceable or barred,

• And whether cost recovery is possible or contractually excluded.

This article offers a comprehensive, technically grounded explanation of the difference. The purpose is not to propose new theory, but to clarify what these concepts mean in real project environments. It explains what tribunals look for, what delay analysts must prove, and what Engineers must assess under FIDIC Clauses 3.5 and 3.7.

We begin by defining these two terms in their technical and legal sense, before exploring their evidentiary consequences.

2.0 Defining Concurrent Delay

Concurrent delay arises when two or more delay events occur within the same time frame, are attributable to different parties, and each independently delays project completion by affecting the critical path.

The Society of Construction Law (SCL) defines true concurrency as:

“The occurrence of two or more delay events at the same time, one an Employer risk event and the other a Contractor risk event, and the effects of which are felt at the same time”1.

Under this definition, concurrency is not established merely by overlap. It requires both delays to:

• Be active on the critical path (i.e. they must drive the project completion date),

• Be independent of each other (i.e. not one causing the other), and

• Be causative — meaning each delay must, on its own, have been sufficient to delay the project2.

The AACE International Recommended Practice 29R-03 reinforces this, stating:

“Concurrent delay occurs when two or more delays occur during the same time period, and the delays independently affect the completion or some other key milestone”3.

2.1 Practical Example of True Concurrency

Consider the following scenario:

• The Employer delays approval of revised HVAC coordination drawings for the upper floors by 10 working days.

• Simultaneously, the Contractor’s subcontractor fails to deliver façade cladding materials, causing a separate 10-day delay.

If both events:

• Occur on separate critical paths (e.g. mechanical systems vs. envelope closure),

• Have zero float, and

• Each would have independently delayed project completion,

…then true concurrency exists.

In such cases, the contractor would normally be entitled to an EOT (for time), but not to prolongation costs, a position endorsed by English law courts and many arbitrators, following what is often referred to as the Malmaison principle3.

2.2 Key Technical Thresholds for True Concurrency

A delay cannot be labelled as concurrent unless:

• It would, in isolation, have delayed completion (not just progress),

• It occurs on a path that has no float,

• It is not conditional on the other delay (i.e. it is independently effective), and

• It is not trivial in relation to the dominant cause of delay5.

In Adyard v SD Marine, the court ruled against the contractor because although Employer delays overlapped with Contractor delays, only the Contractor delays had the capacity to delay project completion6.

Tribunal position: Overlap ≠ concurrency.

Concurrency = independent causative effect.

2.3 Why True Concurrency Is Rare

Despite its frequent use in claims language, true concurrency is uncommon in real projects. Most overlapping delays:

• Occur on different paths, one of which is non-critical;

• Involve delays of different magnitude (e.g. 2 days vs 25 days);

• Or are linked causally (e.g. a Contractor delay is caused by a late Employer instruction).

Pickavance describes this misclassification as “labelled concurrency”, a strategic argument rather than a proven fact7.

Unless the delay is independently critical and causative, it should not be classified as concurrent, no matter how close the dates are.

2.4 Summary

To establish true concurrency in a FIDIC context, the following must be satisfied:

• The delays must be caused by different parties;

• Each must affect the completion date;

• Each must be independently critical and not absorbed by float;

• There must be robust schedule analysis, not just overlapping dates.

Misapplying this label, as we will explore next, has become a frequent basis for unjustified claim denials, often corrected at tribunal level.

3. Understanding Parallel Delay — And Why It’s Not Concurrency

While the term “concurrent delay” has a recognised technical and legal definition, the term parallel delay is used more informally, often as a practical label in project management and dispute narratives.

It generally refers to two or more overlapping delays that do not meet the criteria for concurrency, particularly where:

• Only one of the delays is on the critical path;

• Both delays are caused by the same party;

• Or the overlapping delays do not independently affect completion.

The key distinction lies in the causative weight and party responsibility.

As noted in AACE RP29R-03, overlapping delays caused by the same party — even if they occur simultaneously, do not qualify as concurrency for entitlement purposes8. Likewise, in many tribunal proceedings, delays referred to as “parallel” are simply internal sequencing issues or non-critical overlaps.

3.1 Examples of Parallel Delay

Example 1: Internal Contractor Delay

• A contractor experiences a delay in steel reinforcement delivery and a simultaneous manpower shortage.

• Both occur within the same week, but both are internal.

• This is a parallel Contractor delay, no EOT entitlement, full Contractor liability.

Example 2: Employer Variation + Contractor Disruption

• An Employer delay (e.g. late instruction) and a Contractor disruption (e.g. workshop shutdown) occur during the same period.

• Only the Employer delay affects the critical path.

• The Contractor disruption affects a float path.

• This is not concurrency, but parallel delay, EOT may still apply in full.

Example 3: Employer issues two variation orders

• One affects the façade; the other effects basement MEP works.

• Both cause delay but in different areas.

• These are parallel Employer delays. The Contractor may still claim the combined effect, but these are not concurrent unless both hit critical paths.

3.2 Tribunal Treatment of Parallel Delay

Courts and arbitrators are generally not interested in overlapping delay dates unless causation is proven. In Walter Lilly v Giles, the judge rejected the employer's argument that the contractor’s minor delay negated the EOT claim. The court found the contractor’s delay was not on the critical path, and therefore non-causative4.

In Adyard v SD Marine, the court emphasised the “but-for” test — if the Employer’s delay would not have altered the completion date in light of dominant Contractor delays, the Contractor is not entitled to time6.

Tribunals do not treat parallel delay as a trigger for “time but no money.” That principle — derived from Malmaison — applies only to true concurrent critical delays3.

3.3 Summary: Why Parallel Delay Should Not Be Labelled Concurrency

The distinction is vital. Labelling a parallel delay as concurrent can:

• Result in an unfair EOT rejection,

• Trigger costly disputes,

• Undermine good-faith contract administration under FIDIC Clause 3.5 or 3.7.

Tribunals expect parties to prove concurrency with forensic clarity, not to assert it loosely based on overlapping Gantt bars or blended dates.

“Concurrency must be established on the facts, not on programme coincidence.”9

4. The Role of Float — And Why It Matters for Concurrency

In delay analysis under FIDIC and other standard forms, float is a central concept — and its treatment often determines whether a delay qualifies as concurrent, parallel, or causally irrelevant.

4.1 What Is Float?

Float (also called total float) is the amount of time an activity can be delayed without affecting the overall project completion date. In critical path method (CPM) scheduling, activities on the critical path have zero float, meaning any delay will push completion.

Delays that consume float but do not affect the critical path are often dismissed in EOT analysis, unless they ultimately affect completion.

4.2 Float and Concurrency

Concurrency cannot exist unless both delays impact the critical path.

The SCL Protocol states:

“Where one event would delay the works whether or not the other occurred, there is no concurrency”1.

In practice, many Contractor delays alleged to be concurrent are in fact float-consuming, not critical. When the Employer's delay is critical and the Contractor's is not, concurrency does not apply, even if dates overlap.

4.3 Float Ownership Debate

A common dispute is: who owns the float?

• Contractor’s position: The float was built into the programme and is under their control.

• Employer’s position: Float belongs to the project; it's a shared buffer.

• Tribunal consensus: In the absence of explicit contract terms, float is a project resource7.

The delay that consumes the float first is generally considered the effective cause of delay7.

As Pickavance explains:

“Float is best treated as a consumable — and causation flows to the event that burns it first”7.

4.4 Tribunal Practice

In City Inn v Shepherd6, the court apportioned delay between Employer and Contractor based on criticality and impact, not simply on presence. The court found that Contractor delays were non-critical in several cases, hence did not negate EOT entitlement.

In Adyard, the contractor’s delays, while overlapping, did not consume float after the Employer’s delay. As such, the court concluded that the Employer’s delay had no effective impact on the overall completion6.

4.5 Summary: Float Filters Causation

Float is not just a buffer — it is a causation filter.

• If a Contractor delay is absorbed by float, it does not cause delay to completion.

• If the Employer delay consumes float and causes critical delay, it may trigger EOT — regardless of minor overlapping Contractor delays.

Unless both delays are critical and causative, they are not concurrent.

5. Tribunal Standards for Causation: But-For, Dominant Cause, and Evidentiary Thresholds

No tribunal will grant an EOT, apportion delay, or allow a prolongation claim simply because two delays happened at the same time. What matters is causation — and tribunals apply established legal and forensic standards to determine which delay actually drove the completion date.

The test is not timing.

It is effect.

5.1 The “But-For” Test

The but-for test is the dominant test in most common law systems:

Would the project have been delayed but for the Contractor’s delay?Would it have still been delayed but for the Employer’s delay?

If the Contractor’s delay would not have affected completion in the absence of the Employer delay, then the Employer delay is causative, and the Contractor is entitled to an EOT — even if both delays appear in the same window.

This test was decisive in the UAE case Adyard Abu Dhabi v SD Marine5. The contractor claimed EOT for Employer-caused variations but had its own delays. The court applied the but-for test and found that the contractor’s delay was the dominant factor — the Employer delay made no difference to completion.

5.2 Dominant Cause

Where the but-for test results in both delays appearing to impact completion, tribunals often look for the dominant cause — the delay with greater practical effect.

In Walter Lilly v Giles4, the Employer argued that Contractor inefficiencies nullified the EOT entitlement. However, the court found the Employer’s delays were dominant — they were on the critical path and controlled the actual handover date. EOT was granted.

“Where both causes are critical, the one with the stronger causative potency usually prevails.” – Uff QC8

5.3 Equal Causation – Malmaison Principle

In cases where both Employer and Contractor delays:

• Are independently critical,

• Happen during the same period,

• And have equal causative potency,

…the courts typically apply the Malmaison principle3:

The Contractor is entitled to time, but not money.

This approach reflects a fairness balance:

• The Contractor gets protection from LDs (since the Employer caused part of the delay).

• The Employer avoids having to pay prolongation costs (since the Contractor was also at fault).

5.4 Apportionment Approach – City Inn

In Scotland, the court in City Inn v Shepherd6 adopted a flexible apportionment test.

The court found that:

• There were multiple critical delays by both parties,

• None could be considered clearly dominant,

• And apportionment was the fairest solution.

The court allowed a pro-rata adjustment to the EOT, reducing it by the proportion attributable to Contractor fault.

While apportionment is not accepted under English law for EOT purposes, it is increasingly used by arbitral tribunals in international projects and under UAE Civil Code Articles 290–29110.

5.5 Evidentiary Thresholds

To satisfy a tribunal, a Contractor (or delay expert) must present:

• Clear CPM logic with updated as-built data,

• Narrative proof of independent causation,

• Float analysis showing absorption or exhaustion,

• And ideally, a Windows Analysis or Time Impact Analysis that isolates each delay.

Simply showing that delays occurred concurrently in time is not sufficient. As stated in Walter Lilly:

“The mere fact of overlap does not create concurrency unless both events delayed completion.”4

6. UAE Civil Law Perspective: Articles and Concurrency Treatment

While FIDIC-based projects in the UAE typically follow international standards when it comes to delay classification, the local legal framework under the UAE Civil Code adds a layer of nuance, particularly when both parties contribute to delay. Unlike common law jurisdictions, the UAE Civil Code does not formally define concurrent delay, but it does contain general principles that apply when multiple causes contribute to harm.

6.1 Article 290: Shared Fault and Contributory Negligence

Article 290 of the UAE Civil Code states:

“If the person suffering harm contributed by his own fault to bring it about, the compensation shall be reduced proportionately.”10

This mirrors the principle of contributory negligence in English law. In the context of delay claims, if the Contractor’s fault contributed to the delay (e.g. by manpower underperformance or subcontractor inefficiency), the tribunal may reduce or eliminate compensation, even if the Employer was also at fault.

But reduction does not mean elimination.

If the Employer’s delay also caused the delay to completion, then time relief, i.e. EOT, may still be justified.

6.2 Article 291: Multiple Party Liability

Article 291 provides further guidance:

“If several persons are responsible for the harm, each shall be liable in proportion to his share of the fault.”10 

This article enables the apportionment of delay liability where both Employer and Contractor contribute to a delay window. The court (or tribunal) may allocate percentages of responsibility, depending on which delay was more critical, longer in duration, or first in sequence.

This is highly relevant in cases where:

• Both parties caused delays that overlapped, but

• Neither party’s delay was clearly dominant, and

• The Engineer rejected the EOT entirely, citing “concurrency.”

6.3 Application in Case Law – Dubai Cassation Court 161/2021

In a landmark ruling, the Dubai Cassation Court affirmed that:

• A Contractor can still receive an EOT even if a Contractor-caused delay occurred during the same period;

• If the Employer’s delay independently affected project completion, time relief applies;

• The Contractor’s delay would only bar compensation if it was more causative, or if no proof was presented to show Employer impact10.

This judgment is significant because:

• It aligns with international best practice (Malmaison, SCL Protocol),

• It rejects the idea that any Contractor delay negates an EOT, and

• It reinforces the importance of proving critical path impact.

6.4 Civil Law Trend: Fairness Over Formula

Tribunals seated in the UAE have increasingly adopted a hybrid approach:

• Applying causation principles like the but-for test, and

• Falling back on apportionment where causation is unclear.

This reflects a broader civil law ethos: the aim is to ensure fair risk sharing, not strict liability based on isolated delay snapshots.

7. Practical Guidance: For Contractors, Engineers, and Delay Experts

The successful resolution of concurrency disputes does not hinge on who tells the best story, but on who presents the most credible and causally precise evidence.

Whether you are a Contractor preparing an EOT claim, an Engineer tasked with determining it under FIDIC Clause 3.5 or 3.7, or an independent delay expert advising on arbitration, the treatment of concurrency must meet strict evidentiary thresholds.

7.1 For Contractors

A common mistake in Contractor submissions is referring to overlapping delays as “concurrent” without proving that:

• The Employer’s delay was critical,

• The Contractor’s delay was also critical,

• And both delays were independent and simultaneous.

To avoid this:

Use structured delay analysis:

• Apply Window Analysis or Time Impact Analysis (TIA).

• Show how the Employer’s delay extended the critical path.

• Use updated as-built data (not only baseline comparisons).

• Demonstrate that the Contractor delay either:

  • Was on a different path (i.e. not critical), or

  • Was already absorbed by float.

Prove entitlement through:

• Narrative causation — describing each event’s timing and impact.

• Programme overlays — showing when and how each delay occurred.

• Substantiated evidence — email trails, instruction logs, RFIs, and technical logs.

⚠️ Do not label a delay “concurrent” unless it satisfies the SCL or AACE definition — tribunals will expect you to prove it1,2.

7.2 For Engineers

Engineers assessing concurrency under FIDIC must act with caution and transparency.

FIDIC 1999 Clause 3.5 and FIDIC 2017 Clause 3.7 require: “A fair determination, taking due regard of all relevant circumstances.”

This means:

• Analyse both Employer and Contractor delays independently.

• Determine whether each affected the critical path.

• Check if one delay consumed float; if yes, the other may not be concurrent.

• Avoid blanket statements like “both parties were delayed — claim rejected.”

Best practice for Engineers:

• Document reasons for determination (or rejection),

• Refer to both parties’ positions,

• Avoid using concurrency as a “default shield” against Employer-caused delays.

A tribunal is unlikely to give weight to a determination that simply denies relief without identifying the effective cause of delay4, 6.

7.3 For Delay Experts

As an independent expert preparing a report or opinion, you must:

• Explain your concurrency methodology (e.g. SCL-based, AACE-based),

• Declare your causation model (but-for, dominant cause, apportionment),

• Reference accepted case law (Malmaison, Walter Lilly, Adyard, City Inn),

• Explain clearly whether the delays were:

  • Concurrent and critical,

  • Parallel but non-critical,

  • Or float-absorbing.

Include:

• Critical path schedules with delay overlays,

• As-built vs as-planned comparisons with daily granularity,

• Float consumption charts (if relevant),

• A clear conclusion on who caused the delay to completion — and whether concurrency exists.

Experts are expected to educate the tribunal — not merely support a party's position8,9.

8. Tailpiece: Clarity, Causation, and Contractual Discipline

Concurrency is not about two delays existing on the same page of the programme.It is about two parties, each causing delay, at the same time, each delay independently affecting project completion.

This distinction is more than contractual — it is evidentiary.

It forms the basis on which:

• Engineers issue fair determinations under FIDIC Clause 3.5 or 3.7,

• Contractors assert or defend their EOT entitlement,

• Experts formulate delay opinions,

• And tribunals weigh liability and assess damages.

Misusing the term “concurrency” to reject or dilute valid claims undermines fair contract administration and exposes projects to protracted disputes.

Similarly, attempting to disguise float-absorbed or non-critical delays as “concurrent” risks a full denial of relief in arbitration.

The way forward lies in:

• Defining concurrency precisely,

• Demonstrating causation clearly,

• And respecting the forensic principles recognised across tribunal rulings and industry protocols.

Tribunals expect proof, not assertion.

Engineers are required to determine, not defer.

Contractors must distinguish real concurrency from overlap.

Because in the end, the delay that moved the completion date is the one that matters.

And the party who caused it must bear the consequence, whether in time, in cost, or both.

[1] Society of Construction Law, Delay and Disruption Protocol, 2nd Edition (2017).

[2] AACE International, Recommended Practice No. 29R-03: Forensic Schedule Analysis (2011).

[3] Henry Boot Construction v Malmaison Hotel [1999] 70 Con LR 32.

[4] Walter Lilly & Co Ltd v Giles & DMW Developments Ltd [2012] EWHC 1773 (TCC).

[5] Adyard Abu Dhabi v SD Marine Services [2011] EWHC 848 (Comm).

[6] City Inn Ltd v Shepherd Construction Ltd [2010] CSIH 68.

[7] Pickavance, K., Delay and Disruption in Construction Contracts, 4th Ed., Informa Law, 2010.

[8] Uff QC, J., Construction Law, 11th Ed., Sweet & Maxwell, 2022.

[9] Knowles, R., 200 Contractual Problems and Their Solutions, 2nd Ed., Wiley-Blackwell, 2005.

[10] UAE Civil Code, Federal Law No. 5 of 1985, Articles 290–291.