Back To Blog
Construction Data

AI & ADCMS: Clarifying Compensable Delays

May 06, 2024

Avoiding the wheel of fortune with claims

For anyone who has ever experienced a project that went sideways in a significant way, enough can’t be said about what the value of good documentation would have been, had it been readily searchable, available, and viable for working cooperatively to see where the issues truly lay. Good documentation would have supported informed decision-making about how to react (or proact) and maybe even show that the entire situation was avoidable. Or, at least, that the situation that could have been more effectively managed to a better outcome.

What is known, is that claims for delay and compensable delay, are the single, largest contributing factor to infrastructure projects running over budget by significant sums, especially when compared to those costs incurred during other stages of the project lifecycle. Anyone, who’s ever been part of a claim in the seven-figure plus range of dollars, knows this all too well. And, shockingly, the frequency of these types of outcomes is only on the rise. Yet, it takes a keen understanding of the process behind these claims, the way that adoption of an Advanced Digital Construction Management System (ADCMS) for the project can not only unite all parties behind a single source of truth, but can also act as a preemptive arbiter of issues that typically are the root cause of such claims, to understand it’s true value and see it’s true potential. When you add in the emerging capabilities of AI and the ability of AI tools to quickly process vast sums of information, the ability to leverage good documentation into a more efficient and effective project management process increases tenfold.

As an example of the current status of this type of situation, we’ll use the case of a contractor who bids for a roadway reconstruction project designed by a local town. During the early course of construction, the contractor realizes that the utility information within the project isn’t reliable. The net effect is that certain assumptions of how the project was to be performed during the bid process, can no longer be performed that way. Instead, piecemeal operations mean the contractor has now taken exponentially longer to complete a critical path item which has set the project back months, despite their best efforts to accelerate. Ultimately, the contractor is forced to make a compensable delay claim for being forced to work out of sequence and in smaller parts culminating in a financial impact for costs not covered by the unit basis of work items, such as additional mobilizations, additional material costs, and additional fuel and manpower including losses for lost efficiency.

In most cases, with a project like this, the contractor is required to submit an initial schedule and bi-weekly schedule. During the course of the project, those schedules rarely are reflected in an update schedule which captures the changes or progress. It isn’t until it’s apparent that the project is in trouble that everyone retreats to their corners, and the contractor comes out with the opening salvo of a claim, backed by documentation supplied by their side alone. Documentation consisting of a now updated schedule based on recollection and internal superintendent notations, loose timelines, and inadequate data. Once the inspection team has the opportunity to rigorously review the documentation against their own in an attempt to refute and obfuscate any of the details, the parties almost universally end up in a room with their respective counsels where they agree to a scenario everyone can live with, at a price tag that’s affordable. But what if it didn’t have to be this way?

Enter adoption of an ADCMS like Appia - created before the rise of ADCMS terminology, but designed to perform all the same functions and more. While on the outside looking in, it’s plain to see what the system is designed to do, and what it can do - daily work report capture, photos, document controls, item placements, payments, change orders, and more. But what is often overlooked is the value of the data points intrinsic to each of those things that supply value in ways not overtly apparent or part of their design. Consider for a moment that each item mentioned on a daily work report, including when and how much of a pay item is placed, is categorized with a date. Dates that dictate when work started and when it was finished. Also described are aspects of the work like how much manpower and equipment were used. The adage, “time is money” is never truer than in construction projects. And when terms like “digital as-built” are used, oftentimes we think of those being a singular entity, like a digital model, or a set of electronic plans with markups. But in reality, “digital as-builts” should strive to encompass not just what was built versus designed, but how long it took versus proposed, and how much it cost versus bid, and all the little detailed changes along the way.

In the same vein, an ADCMS grants the advantage of strict data control - a crucial aspect of eventually using that data for AI, which thrives on the principle of good data in, good data out. If your organization doesn’t have consistent data, AI won’t be capable of providing reliable information based on that data. In Appia, for example, daily reports can’t be modified except by the person who created it. There’s an indisputable integrity that comes with the audit trails created in these systems versus those in a cluttered filing cabinet. Without clear categorization and definition of project data, there’s no point in leveraging AI for greater insight.

So, imagine an ADCMS, where all those subsets of data points like when, where, and how much could be mined and combined using AI as part of the ongoing work in a way that allowed the inspection forces and the contractor to see not just the number of “working days elapsed versus dollars spent,” but gave keener insight into “time elapsed on items versus proposed” and the dollar cost associated on a more granular level? What if, in the example above, the contractor and inspectors could see in near-real time what the impact of broken strings of work days of a particular pay item was? Or the under or overstaffing of items on the critical path toward the acceleration of a schedule? What if a contractor deciding to change course could yield with certainty of schedule impact a change to the critical path, all based and pulled from the existing and daily information of an ADCMS? Suddenly, parties to the project would not only be empowered to make better informed decisions at the time of construction, but everyone would be working from the same single source of truth; just as we are chasing for design, we’d be chasing for time and funding.

The proposed net effect would result in reduced confrontation of parties, reduced time to identify issues, correct course, substantiate claims, reduce claim dollar amounts, affect better coordination and communication between parties, and set projects on a course toward leveraging solutions rather than setting up adversarial positions. In a market as dynamic as today’s construction industry where technology adoption, labor shortages, and material and supply chain disruptions are all having significant impact, striving to better deliver projects through cooperative means and thoughtful adoption of AI can’t be of greater importance.

At Infotech, we are currently in the vision-casting stage of exploring how AI can be incorporated into our Appia platform to support the use cases mentioned above. If you have ideas or want to learn more about how Appia supports comprehensive record-keeping and reporting on infrastructure projects, feel free tocontact us.


Adam F. Dawidowicz, CCM
Senior Account Manager
Adam is a Senior Account Manager with a proven track record of growth, boasting 24+ years experience in construction and 15+ years specific to construction inspection. Adam is a subject matter expert in the fields of construction, project management, land surveying, and utilization of technology to further streamline and bring meaningful and valuable savings and results to owners and end-users of data.