Best Use of Technology: Artificial Intelligence

This category recognises teams using artificial intelligence-driven technology to improve project planning or delivery. This may be in infrastructure or on a building project and can cover areas such as improving the design, management or construction through artificial intelligence or being used to aid clients to better plan infrastructure or construction investment.

Aiforsite

Wendy, the AI Assistant for the Built Environment

Wendy, the AIFORSITE's , AI Assistant for the Built Environment is “THE game changer for the industry.” She is there to provide you with the information/data/details/report YOU NEED. When YOU NEED it, in the FORMAT YOU NEED it, in REAL TIME - you could be a designer ensuring your specifications do not include any Red List Materials, or wish to have a BREEAM Assessment done over night on your current version, or an operative in the Basement Plant, requiring the latest pipework drawings, you can ask Wendy in your own mother language for the latest details or a Project Manager who wants to know who is where and are their skillset of everyone on his project, or A Managing Director needing the latest H&SQE/Cost and program status on all his/her Portfolio as to 5pm last night, NOT 3/4 weeks ago, Then Wendy is the AI you need in your corner.

ALICE Technologies and Implenia

AI-powered construction optioneering software

Implenia is exploring locations for a production facility to undertake serial production of concrete foundations for floating offshore wind farms. It enlisted support from ALICE Technologies, the provider of the world’s first AI-powered construction optioneering software, to optimise serial production to ensure that the most efficient outputs can be achieved. ALICE is being used to answer: · What the most efficient site layout is for constructing and moving the foundations · Whether one or two launching corridors is optimum for increased productivity on-site · How on-site resilience can be improved Implenia needs to find a site that will yield up to one gigawatt of energy generation per year once operational. Implenia has found that planning with ALICE is ten times faster than traditional methods. Through optimisation with the AI software, it has been able to boost potential wind turbine foundation production levels considerably.

Efestos Hub

Multimodal-AI Technology

Efestos Hub's multimodal-AI technology aims to strongly increase the flux of available reclaimed steel from donor assets to meet steel demand and reduce carbon emissions from primary steel production. The innovation is based on a combination of machine (ML) and deep learning (DL) algorithms and a novel approach of receiving data via EH’s Designer’s Optimisation Matcher tool (already at testing by users) that creates current and future demand (hence steel worth can be estimated) as well as increases reuse rates (finds suitable reclaimed steel from stock and donors). The proposed solution quantifies the factors affecting the reuse of building structural steel elements and develops a predictive evaluation model to determine the technical reusability of the load bearing steel beams and columns, using machine learning techniques. The novelty lies in the multiple sources of data, and nature of multimodal-AI that utilises multi-dimensional data types to analyse the given complex problem comprehensively.

Free4m Lighting

Free4m Lighting Ltd recently completed its Highway Assets LIDAR/Optical Survey and Ai projects providing clients with tailored public realm asset inventories formatted for uploading to existing client systems. Free4m drove the entire length of their local authority client highway network with walking survey to all footways producing citywide digital twins using LIDAR and high-resolution cameras. In-house authored Artificial Intelligence algorithms were written using R-CNN Ai modellers, tailored to client asset requirements capturing over twenty-five asset types and locations including sizes and offsets of lighting column attachments. Free4m then used the lighting column and attachment data gained from the survey to structurally assess the entire lighting column stock, fourteen thousand columns, with actual loads and strengths versus theoretical leading to a better understanding of actual structural utilisation and column residual life and better knowledge of asset condition. This in turn has impacted the client’s future spend profile and budgeting.

GScan

Founded in 2018. GScan’s pioneering technology pinpoints defects removing doubt by providing infrastructure asset owners and operators with reliable, accurate data about their asset’s integrity to 1mm accuracy. This can save unnecessary decommissioning and enable far greater clarity on the assets status and reduce costs, bringing value to all aspects of asset management. The climate crisis and changing regulations mean it is more important than ever to maintain critical infrastructure and save construction and reduce carbon outputs where possible. Experts agree that the “greenest” assets are the ones already built. GScan’s technology allows customers to make informed decisions about how to optimise reconstruction effort, capital expenditure, and environmental impact. Sustaining existing infrastructure potentially saves up to 60% in construction carbon emissions, saving millions of tonnes of carbon, and provides significant reductions in reconstruction costs.

Mott MacDonald

Machine learning applied on a water pipeline

Machine Learning (ML) was applied on the a project to build a Water Pipeline project which aims to provide sustainable water supply for 10 million people. This technique was applied to derive new geotechnical parameters that are required for the assessment of liquefaction hazard at the location of the proposed Water Pipeline. This work showed that Machine Learning models can be used in engineering resiliency projects to obtain higher accuracy estimation of engineering parameters. The accuracy led through the use of the novel technique avoided over-conservatism in design, ultimately reduced the quantities of engineering material/fill required and improved the sustainability of the project.

Arup

The M5 Metro project, Copenhagen

The M5 Metro project in Copenhagen has utilised artificial intelligence, specifically large language models, through an application called ProjectGPT to help the integrated delivery team find the information they need, when they need it . Major infrastructure projects generate massive amounts of documentation at each stage of development, and it is a chore for people delivering the project to find the information they need. A user can ask ProjectGPT natural-language questions about the project via a chatbot-style interface and receive project specific responses along with a citation, enabling the user to check the validity of the response and access the primary source(s). Looking beyond information retrieval, there is opportunity for using this type of technology to draft base responses to RFIs or summarise content for different stakeholders.

Safety Shield Global

AI safety innovations

Safety Shield Global is an award-winning technology solutions manufacturer developing cutting-edge artificial intelligence (AI) safety innovations. We are leading the development of AI object recognition, collision avoidance and data capture for plant machinery. Construction workers face exceptionally high levels of risk in an already dangerous sector. Our camera-based collision avoidance system uses AI to identify people who are in the danger zone of moving plant and machinery. Trained to filter out all other objects, the system only detects the human form, acting as a third eye for the operator – sounding an alert if a collision is imminent. Our aim is to save lives.

Suction Excavation

Safety exclusion zone with AI detection

Suction Excavation UK's SEUKFHOSSFILED safety exclusion zone with AI detection is a trailblazing innovation that enhances safety, efficiency, and operational effectiveness in vacuum excavation. This world-first technology reflects SEUK’s dedication to pioneering advancements and our commitment to setting new industry benchmarks. The potential adoption of this AI-driven safety solution as a standard practice signifies a monumental shift towards safer and more efficient excavation practices. In summary, SEUK’s innovative approach and significant contributions to the industry make it a deserving candidate for the 2024 TechFest Award. By leveraging AI technology to improve project planning and delivery, SEUK is not only enhancing safety but also driving the industry forward, exemplifying the transformative impact of technological innovation.

WSP

Planning for an unknown future: The adaptive planning algorithm

In the world of fixed assets, decisions are becoming increasingly complex and the uncertainties around what the future may bring (in terms of changes in climate, technology, population needs and political winds) are growing exponentially. Decisions increasingly also have to consider factors such as carbon and biodiversity which are difficult to accurately quantify and around which there are differing views on relative valuation. The Adaptive Planning algorithm enables all of that complexity and uncertainty to be codified and a set of concrete decisions optimised to create a clear, evidenced and justified investment plan to follow. The Algorithm also plans in adaptation points, enabling seamless pivoting of the plan to respond to emerging information about how the future is actually going to look. Through this it is possible to build a robust and defendable investment plan even in the face of high uncertainty and provide the confidence to proceed.