Application of the AI-Enabled Unmanned Aerial Systems in Resolving Highway Construction Claims

Claims are among the most concerning disturbances in the construction sector universally. Simply speaking, they arise from delayed procedures in answering inquiries, solving conflicts, and clarifying vague data to owners. Their development and delayed solution could pose many kinds of threats, like complex court disputes from simple arguments and controversies, contributing to arduous obstacles to complete the whole project progress flexibly. They consume much energy, time, financial resources, and focus from project managers, contractors, seniors, and consultants. Swift response to solve and handle such issues is crucial as their lengthy prevalence would create further problems, resulting in more complexity and common enterprise concerns of cost overruns and delay of the submission schedule. Even the latest building information modeling (BIM), project management (PM) techniques, risk management (RM), and other innovative PM advancements may face different hurdles to provide comprehensive handling. But trials and research and development (R&D) are still conducted, aiming to configure holistic strategies of such challenging affairs. Unmanned aerial vehicles (UAVs) have been innovated and limitedly practiced in specific military uses. Nonetheless, with the rapid boom of digitalization and its involvement in every field, UAV applications have changed into more beneficial, safe, cost effective, and practical uses. Despite their common limitations of public privacy and safety penetration, the scale of positive awareness, interest, and knowledge in many fields, like construction, are overcoming these restrictions. Additional proof is provided, day after day, of their relevance and contributions to conduct many construction activities efficiently and profitably through several UAV-supervised piloting and operating trials in construction and other disciplines.   In this context, this study aims to supply sufficient evidence-based practices necessitated for many construction stakeholders, on UAV feasibility, practicality, and reliability to take a part in managing construction claims actively. Strictly speaking, their importance in this domain is reflected in conducting high-performance, accurate, swift, flexible, and massive data collection since of their high maneuverability capability to cover and inspect large surface areas, offering an alternative option of routine, risky, technically complex, and lengthy physical inspection processes (PIPs) implemented broadly nowadays. Thus, they can minimize much cost, time, risks, and challenges to perform such activities, especially if some acute weather conditions or natural disasters exist. Also, their contribution can be remarkably realized for performing active, broad inspection and coverage of massive surface areas of vertical construction (VC) projects and utility-scale infrastructure city enterprises. For claim consultants and many project stakeholders, this capability is a significant advantage since it helps accelerate and facilitate the management of time-consuming, complex, and energy-exhausting claims and disputes. “UAVs to manage construction claims” is a very rarely discussed subject. It is considered a significant knowledge gap, which needs to be bridged. The current study relies on one secondary data collection (systematic review and meta-analysis of the current knowledge body). Besides, it utilizes four primary data collection approaches, namely cross-sectional quantitative data collection and analysis by online survey questionnaires, covering expert construction engineers in Illinois Chicago, cracking severity analysis by artificial intelligence (AI) algorithms, certainly high-performance deep learning (DL) frameworks, and holistic examination of different local, public, and federal legislative frameworks to facilitate and enable active UAV application in many construction activities. A framework containing recommended amended acts is formulated to allow these regulations to support UAVs in recording helpful data for swift and flexible court claim or complex dispute management. The overall outcomes from this research revealed a collection of important statistical facts and imperative data. Firstly, the secondary data collection, from the systematic review and meta-analysis, uncovered rapid growth of UAV involvement in diverse promising applications and contributory utilizations, helping supply enhanced accuracy, high-quality data, and savings of energy, time, labor effort, and cost resources. The literature reported that UAVs can make beneficial photogrammetry tasks, massive and rapid aerial inspection of various types of construction sites, which are difficult to reach. UAVs can develop daily progress reports, take precise visual image or video records of critical data, share in many influential tasks of PM and RM in normal conditions or at natural disasters. They can provide critical risk estimation data for risk decision makers and responsible authorities to save cost, equipment, and human resources. The estimated Cronbach's Alpha coefficient of the survey dimensions was 0.885, indicating higher internal consistency among the items, exceeding the recommended threshold of 0.7. Approximately, through all survey articles, more than 80% of the overall surveyed engineers (135 individuals) have affirmed the relevance, contributions, advantages, and profitability of UAVs in construction to manage construction claims flexibly, efficiently, and functionally. In-depth legislation analysis of UAV-related laws, notably UAV 620 ILCS 5/42.1, UAV 725 ILCS 167, Federal Aviation Administration (FAA)/ Part 107 Law, Public Act 100-0735, and Public Act 725 ILCS 167 has identified many significant gaps, application voids, and affairs that need to be carefully taken into consideration by UAV supervisors, construction stakeholders, and aviation law makers to ease UAV integration in construction for many advantageous applications, specifically for practical construction claim and dispute management (CC&DM) and alternative dispute resolution (ADR). Concerning the critical AI classification outcomes, it was found that convolutional neural networks (CNN) DL framework has enabled a fully automated process of high-way bridge cracking severity prediction existing in different walls and structures, which is considered technically very complex, costly, and time-consuming. CNN provided an accuracy, reliability, and effectiveness ratios of over 95%, identifying ‘critical’ from ‘noncritical’ cracks. The CNN DL recognition process is aligned with the American Association of State Highway and Transportation Officials (AASHTO) code.