BAM Nuttall is one of Europe's largest contractors and operates a large number of complex construction projects. It developed the Learning Camera using computer vision to improve health and safety monitoring. Computer vision uses onsite cameras and machine learning to train computers to accurately identify and classify objects and then react to what they see. It now wants to apply this technology to streamline project management and progress reporting in the supply chain and create a single source of truth about each partners' contribution to a project.

In supply chain, where there are many partners and contractors, it can be hard to reach consensus about things like estimation, progress, assurance, and even responsibility for delays or errors. And as there is no objective monitoring of activity onsite or effort employed, reporting to clients and stakeholders can vary depending on different partners. This can lead to blame being passed along the supply chain, which can slow projects down or require extra time and money to settle contractual disputes.

One Source of Truth automates data collection around project management, scheduling and effort employed, both during and post-build. It can simultaneously streamline estimating, quality assurance, productivity planning, project management, contract administration and carbon accounting. Contractors, clients and supply chain no longer have to rely on subjective feedback about progress or conditions onsite. Instead the complete digitised records from One Source of Truth gives total visibility and will improve the accuracy and objectivity of information flow. In doing so, it can increase project productivity and quality, and demonstrate the value of Modern Methods of Construction to be able to win future work.

The success of The Learning Camera, and the increased demand to digitise construction records for accountability and traceability, gave BAM Nuttall the impetus to apply its cutting-edge digital technologies (including computer vision, Internet of Things, machine learning and edge computing) to improve the automation, transparency and longevity of information in the supply chain.

BAM Nuttall applied for funding to explore computer vision technology and engage with partners that could explore potential use cases and gaps in the market. It collaborated with the Building Research Establishment, Cranfield University, Glideology and Werner Homes to pilot real-time automated construction monitoring at Houghton Brook Flood Alleviation Scheme.

Subjective or inconsistent data capture can raise questions about effort employed or responsibility for errors and rework. Settling these disputes can add time and cost, and impact productivity and quality of delivery. By turning to automated, objective data instead, everyone in supply chain has sight of the same single truth, which minimises these kind of disputes or delays.

A number of monitoring technologies were tested on the pilot site at Houghton Brook Flood Alleviation Scheme to help capture the complete data needed to create One Source of Truth. Firstly mobile, power-independent cameras were used to capture time lapse footage of large sections of construction sites autonomously through remote control. These cameras are built into safety helmets and stationed onsite. Other sensor sources such as machine trackers and weather station data have been integrated to provide a rich picture of site activity. Artificial intelligence is then trained to analyse the footage received and automatically monitor progress, track moving parts onsite, such as construction plant and people, and improve site security. Edge-processed personal data masking is used to ensure the footage is collected ethically, addressing privacy concerns, and aligns with GDPR Regulations. The technology feeds the data through to a digital twin and event streams of the construction site - through a bespoke dashboard that acts as a one-stop-shop objective record of site activity. The dashboard delivers insights to the site team about the effort employed by each partner, information about progress and conditions onsite and, importantly, can act as an objective source of information about errors in the build process so they can be traced and determine who needs to rectify them. The technology overlays this to the contract terms between the supply chain to minimise disputes around responsibility for rework and reduce delay and cost associated with contractual issues.

BAM Nutall is the lead on this programme and worked with a number of partners to pilot real-time automated construction monitoring at Houghton Brook Flood Alleviation Scheme. Werner Homes helped with the housing-based implementation. BRE explored the housing potential, future funding potential, and further industry connections. Cranfield University developed four data-driven digital modules, powered by AI, to monitor the asset usage and availability, PPE detection, person-hour and working type of staff and plants, and working progress. And Glideology helped with the hardware and software integration

• BAM Nuttall
• BRE Centre for Innovative Construction Materials
• Glideology
• University of Cranfield
• Werner Homes

Transforming Construction and Innovate UK provided funding, access to other innovative companies and consortia, and support to unlock business development opportunities. Three further pilots are confirmed on large projects in the North and Scotland, that started in May 2021.

One Source of Truth means contractors, clients and supply chain no longer have to rely on subjective feedback. Instead the complete digitised records gives total visibility and will improve the accuracy and objectivity of information flow. In doing so, it can increase project productivity and quality, and demonstrate the value of Modern Methods of Construction to be able to win future work. It also mitigates the delays and cost that can arise from contractual disputes as effort employed and accountability for errors is all tracked and recorded.

One Source of Truth collects personal data in an ethical and legal way, and is compliant with GDPR regulations. This was a key consideration in the development of the technology. The information and insights that it provides support better internal feedback between construction partners and business support functions, resulting in more accurate project management, tendering and project reviews There are less disputes between supply chain partners because of the objectivity it provides about who is responsible for rectifying errors and/or rework.

Mobile, power-independent cameras capture time lapse footage of large sections of construction sites autonomously through remote control. Artificial intelligence is then trained to analyse the footage received and automatically monitor progress, track moving parts onsite, such as construction plant and people, and improve site security. The technology feeds the data through to a digital twin and event streams of the construction site - through a bespoke dashboard that acts as a one-stop-shop objective record of site activity. The dashboard delivers insights to the site team about the effort employed by each partner, information about progress and conditions onsite and, importantly, can act as an objective source of information about errors in the build process so they can be traced and determine who needs to rectify them. The technology overlays this to the contract terms between the supply chain to minimise disputes around responsibility for rework and reduce delay and cost associated with contractual issues.

A number of monitoring technologies were tested on the pilot site at Houghton Brook Flood Alleviation Scheme to help capture the complete data needed to create the One Source of Truth. Three further pilots are confirmed on large projects in the North and Scotland, that started in May 2021.

Productivity
This technology overlays information on errors to the contract teams between the supply chain to minimise disputes around responsibility for rework and reduce delay and cost associated with contractual issues. Improved monitoring allows for health and safety on site to be monitored more closely.

Safety
This project develops a novel practical solution powered by AI to measure PPE compliancy in real-time. Improved monitoring allows for health and safety on site to be monitored more closely. The algorithm has been published in the journal, Sensors.