Data, AI & Construction

Julian Haslam-Jones and Janusz Oczyp of Alvarez & Marsal discuss how data management and technology assisted analytics in construction can enable companies and lead to greater efficiency...
United Arab Emirates Technology
To print this article, all you need is to be registered or login on

Julian Haslam-Jones and Janusz Oczyp of Alvarez & Marsal discuss how data management and technology assisted analytics in construction can enable companies and lead to greater efficiency, innovation and sustainable growth.

The construction sector continues to be at the heart of economic development in the Kingdom of Saudi Arabia (KSA). KSA has embarked on an ambitious journey to diversify its economy, with the construction sector playing a pivotal role in this transformation. The KSA construction market is expected to register a significant growth rate, with projections indicating that the sector will reach a value of over USD 130 billion by 2024. This growth is propelled by multiple giga projects, which are not just transforming the landscape but also setting new standards for managing complex data and leveraging technology in construction disputes.

The construction industry in KSA, and globally, is becoming increasingly data driven. The deployment of technologies such as drones, and artificial intelligence (AI) for predictive analytics and dispute resolution is not the future; it's the present.

The large size of the projects is going to generate huge amounts of data. Navigating the intricate interplay of data, technology, and legal frameworks within the construction sector is challenging.

This article explores how data management and technological advancements are shaping the resolution of construction disputes including the challenges and opportunities that lie ahead.

Current technology

Traditionally the type of data produced during construction or engineering projects included correspondence in the forms of letters and emails, P6 Schedule Data, Drawings and procurement logs, Weekly/Monthly progress reports and site photos.

Common new forms of data include smartphone chats, online contract management communications, digital site imagery, 4D simulations, 3D printing and blockchain. The technology associated with this data is continually evolving. How data is managed and stored is extremely important to the commercial viability of the project, including the outcome and the costs related to any dispute.

Case Study 1 – Time-lapse systems

The operator and contractor employed a third-party to produce a 15- to 20-minute time-lapse video recording changes over the entire project duration. It was an important resource for building up the as-built analysis and interrogating the other party's claim, provided a critical contemporaneous record, and helped resolve the dispute.

Case Study 2 – Drones

The drone was managed by the Contractor. Drone footage and data were submitted as part of monthly application for payment.

Drones are used for multiple purposes on construction sites:

  • Initial Site Survey and Measurement
  • Workflow Monitoring
  • Equipment Tracking
  • Remote Monitoring and Progress Reports
  • Complete Inspection of the Project
  • Generation of Advertising Material
  • Risk and Safety Management

Other Examples

Drone technology is not limited to video and photos. 3D laser scans by drones can be fed into the Building Information Model (BIM).

BIM provides digital models of buildings. There are various levels:

  • 3D includes geographical and structure
  • 4D includes timeline, scheduling, and duration
  • 5D includes cost estimation, budget analysis
  • 6D includes self-sustainable and energy efficiency
  • 7D includes facility management information

3D Laser Scanners analyse real-world objects, onsite surveying, mapping, project inspection and safety. An example is an existing building 3D scanned and fed into BIM to generate a renovation of the building.

4D Simulation is a planning process that links the construction activities represented in the time schedule with 3D models to create a real-time graphical simulation of construction progress.

3D printing takes a 3D model and turns it into a three-dimensional building.

A digital twin creates a building prototype to carry out performance analysis and consider occupants' behaviour and use patterns.

Virtual and Augmented Reality provides an interactive 3D environment, where project blueprints can be transformed into lifeline models. This helps visualise, inspect, and understand projects.

Blockchain Technology (Smart Contracts) is a digital ledger providing a record of all distinct transactions. A construction contract is one large, complicated transaction.

Collection of data

Traditionally, construction contracts bring obligations on parties to collect and present key data points: baseline and progress updates of the schedule, engineering and procurement logs, daily site reports, weekly and monthly progress reports, correspondence logs between parties, and payment applications.

Challenges could prevent successful data sharing, such as using third-party data without consent, exclusive rights over data being exercised by technology providers, and competition and privacy law concerns.

Many project management and data collections software packages manage this data. However, such software is generally designed for a forward-looking delivery process and does not consider forensic requirements.

Challenges with project management tools

Many project management tools are prospective. This presents challenges in analysing construction data in a dispute. eDiscovery and analytics tools assist this type of analysis.

Data and project management systems specialised for construction and engineering projects support a variety of data types and can host large volumes of data. Delivered by either cloud or local based providers, systems are familiar with the data requirements generated during the design and engineering process.

These systems are not designed for forensic data analysis. There is no extensive provision for modern searching requirements or storytelling. There is limited functionality for chat hosting and video and audio search, application of AI technologies and task automation, and inconsistent record keeping.

E-Discovery benefits & technology-assisted solutions

In disputes, effective data management is not only about collecting data, but also about presenting it in a clear manner. The ability to access accurate, well-documented information can significantly contribute to the resolution of disputes through negotiation, mediation, arbitration, or litigation.

Benefits of technology include:

  • The provision of a trusted solution to governments, lawyers, and large corporations.
  • Centralised, end-to-end Electronic Discovery Reference Model (EDRM) without, or with limited need for, third-party tools.
  • Automated workflows, pre-defined dashboards, auto-processing, Optical Character Recognition (OCR) and Searching, Productions functions.
  • Cost effectiveness inventory through data management using repository, review management and inbuilt products.
  • In-built products, such as collect, contracts, machine translation, auto-redaction, PI detect, chat view and the industry-standard Relativity Short Message Format (RSMF), allow the review of smartphone chats in formats similar to an actual phone.
  • Technology Assisted Review (TAR) provides Continuous Active Learning (CAL) assisting with email threading, communication analysis, clustering, conceptual analysis and keyword expansion.
  • AI add-on software, such as aiR, which uses open and generative AI to introduce decision-making, scheduling, disputes prevention or similar.
"AI will significantly increase the ability to create the outline of the case, deposition, and witness summaries, practically without reviewing any documents by humans – while generating the chronology of key events based on initial input from reviewers or lawyers."

Transforming Data Management and Review

Technology-assisted solutions are transforming data management. Two key solutions are TAR/CAL and intelligent voice workflow.


TAR/CAL uses AI to identify and tag potentially discoverable documents, expediting the human review process.

Reviewers code documents (e.g. relevant or not relevant) while in parallel the engine continuously learns from the coding decisions. Documents are continuously resorted for relevance, presenting the most relevant document first. This reduces the number of documents subject to review, saving time and costs.

Audio Transcription – Intelligent Voice Workflow

Voice Workflow employs AI, in the shape of machine learning, natural language processing (NLP), and real-time speech recognition to analyse calls, audio recordings, visual voicemail, and digital conversations. This significantly reduces costs compared with human listening and manual transcription methods. The key features are:

  • Process and identification of audio files in e-Discovery environment.
  • Utilisation of external Intelligent Voice software to perform.
  • Voice activity detection.
  • Automatic speech recognition.
  • Categorisation and topic identification.
  • Transcription of voice to searchable text.
  • Transcribed data, exported from Intelligent Voice Data, fed back to e-Discovery platform and overlayed over original voice data to enrich available information.
  • Extensive and complex searching, jump to topics, animated graphic timeline, and TAR analytics performed as any other documents.

Transforming Review with AI

We are still in the early stages of developing AI; however, it is already transforming the review of documents.

Initiating process for the Generative AI engine and review of key documents

AI can assist with reviewing the context of documents. A prompt iteration experience helps identify responsiveness criteria and key issues. The reviewer can see detailed rationale and citations assisting with understanding and defending results, and also see insights directly in the context of documents and accept coding decisions.

Creating automated timeline of key events and decisions with abbreviated story line and a chronology of events

AI will significantly increase the ability to create the outline of the case, deposition, and witness summaries, practically without reviewing any documents by humans – while generating the chronology of key events based on initial input from reviewers or lawyers. It will also allow the extraction of key decisions from the documents, and attachment to the chronology chart, resulting in comprehensive storytelling of events in a concise manner.

AI APIs, useful for investigations, document reviews and eDiscovery, can provide:

  • Label identification, and
  • OCR of handwritten documents.

Utilising next-level OCR capabilities assists in overcoming the current industry-wide data issues. Text extraction is a notoriously difficult part of data review and administration to get right, mainly due to current technological limitations. Now, it is possible to accurately extract readable text from scanned and non-searchable documents with greater precision.

AI concerns

While AI can bring significant benefits to the industry, including to e-discovery processes, there are also limitations:

  • Bias and Fairness:
    • AI algorithms may inherit biases from historical legal cases or training data, leading to biased document categorisation and search results, which may affect the fairness of legal proceedings.
  • Accuracy and Reliability:
    • AI models, especially machine learning models, may not always produce accurate results, and errors could lead to incomplete or incorrect document classifications, which may lead to inaccuracies in outcomes, potentially affecting legal strategies and decisions.
  • Lack of Explainability:
    • AI models used in e-discovery, especially complex ones, may lack transparency and explainability in their decision-making processes, which can make it challenging for legal professionals to understand and trust AI-generated results.
  • Security and Confidentiality:
    • Analysing sensitive legal information by AI introduces security concerns, including access and sharing information to the Cloud to achieve the full benefit of Generative AI. Local laws may prevent Cloud access, and if permissive, the approach can compromise the confidentiality of legal cases, with ethical and legal implications.
  • Scalability:
    • As the volume of electronic data grows, so do concerns over the scalability of AI solutions in handling massive datasets within reasonable timeframes. Inability to scale may lead to delays and inefficiencies in the e-discovery process.
  • Cost and Resource Intensiveness:
    • Implementing AI can be resource-intensive, requiring investments in technology, training, and infrastructure. Smaller firms or organisations with limited resources may face challenges in adopting and maintaining advanced AI-driven e-discovery tools.
  • Human Oversight:
    • Human oversight is still crucial for ensuring the accuracy and relevance of results. Overreliance on AI may lead to oversights and mistakes in legal proceedings.
  • Legal and Regulatory Compliance:
    • AI tools must comply with various legal and regulatory standards; achieving full compliance can be complex. Non-compliance may result in legal challenges and hinder the admissibility of AI-generated evidence.
  • Limitation on Interpretation of Context:
    • AI models may struggle with understanding the broader context of legal cases, including the intent behind certain communications or the significance of specific actions. Contextual misunderstandings may lead to misinterpretations of evidence.


The advance in technology in the construction industry is generating different types of data. While this additional data can assist with resolving disputes, there are challenges in organising and reviewing such data.

Many solutions within the market assist with overcoming such challenges. It is critical that parties consider how to manage the data at the start of the project, including the ability to forensically review such data.

Increased usage of video data (drones, smartphones), 3 and 4D modelling and modern data types in construction projects will require the industry to employ modern techniques able to address the advancement in technology and complexity of contraction data points. Advancements in the use of AI in the construction industry seems inevitable. The industry should embrace AI not only due to its potential to streamline project timelines and identify and pinpoint delays or key decisions, but also due to its prospective ability to enhance productivity, optimise resource management and even improve safety standards.

By leveraging AI technologies such as predictive analytics, robotics and automated monitoring systems, construction companies can achieve greater efficiency, cost effectiveness, and innovation, ultimately leading to sustainable growth and competitive advantage in an ever-evolving landscape.

The Oath Magazine originally published this article In their April 2024 issue, which can be found here.

Originally Published 10 May 2024

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.

See More Popular Content From

Mondaq uses cookies on this website. By using our website you agree to our use of cookies as set out in our Privacy Policy.

Learn More