A longstanding problem faced by transport authorities is the difficulty Artificial Intelligence (AI) can have in accurately detecting people and objects within images. These AI models cannot be trained due to a lack of data covering different scenarios, which is exacerbated further by poor cameras, dirty lenses and low resolutions making detection almost impossible. Zest's Synthetic Data Solution (SDS) improves detection rates and reduces the likelihood of false positives. It works better for edge cases and bad conditions, such as poor lighting or adverse weather conditions. Our innovative approach solves these problems. We can generate labelled training data faster than traditional techniques. As a result, accurate results are achieved much faster and the cost of training machine learning models is significantly reduced. Our project will investigate the feasibility of applying our SDS to create simulated/synthetic datasets, which in turn will train Transport for London's (TfL) future Smart Station Artificial Intelligence (AI) systems. We also aim to investigate if SDS can generate unbiased datasets to test and mitigate 3rd party models for bias. TfL's Smart Station proof of concept (PoC) in 2022-2023 enabled significant steps forward into the use of AI to analyse live CCTV footage to safeguard customers and staff, enhance their understanding of fare evaders, identify anti-social behaviour, and analyse trends in safety and security related incidents. Analysis of data post PoC completion revealed limitations to training the AI systems, because accurate training data was not available from historical CCTV footage, and no other datasets exist to train the models. This resulted in the need to use people to simulate behaviours on cameras to create data to ensure the AI systems received appropriate datasets for machine learning. Our SDS will ensure people do not need to simulate activities, as these datasets can be generated. To achieve this feasibility study, we aim to collaborate with TfL's Smart Station team, to investigate how our SDS can generate datasets to address the issues raised by this lack of training data. We will look to interview relevant staff groups, undertake workshops to seek information as to how/where our SDS can contribute to the success of future Smart Stations in London. We will identify use-case(s) to generate images with our SDS and look to testing the validity and reliability of datasets we collect. The project will make a significant contribution to the design and planning of AI in the transport sector with our SDS.

This proposal is a **feasibility study** exploring a novel Generative AI approach addressing the critical construction challenge of time and cost overruns caused by re-work. **Project entails the following activities**: * Explore feasibility of developing an AI-enabled design tool as a solution using data from previous projects to analyse problematic projects identifying issues requiring additional oversight and management to avoid re-work and time/cost overruns. * Provide AI expertise in developing/deploying Applied AI solutions for construction industry through collaboration between a party in need of the solution and a party that can develop it. Critical factors limiting improvements in productivity in the construction industry are time and cost overruns due to unforeseen challenges and re-work. This project has the potential to unlock huge productivity improvements. **Areas of focus** Our project focuses on the following themes: **Data Driven Decision Making:** * **_Better project delivery_**: using past project data and industry wide benchmarks to develop estimators for accurate project plans; reducing likelihood of delays/cost overruns. * **_Enhanced safety_**: using data to identify safety risks and proactive mitigating measures. * _**Quality control**_: using data to monitor quality of workmanship on construction projects to meet quality standards and avoid rework. **Design:** * _**Generative design**_: using AI to generate new design options based on set parameters/constraints. * **_Analysing problematic projects_**: using AI to analyse data from previous construction projects to identify patterns and predict future outcomes.Innovations Our project sets the foundations for developing a Generative AI-enabled design tool to analyse problematic projects using data recorded by construction companies; offering valuable insights enables them to take proactive measures, allocate resources effectively, make informed decisions to mitigate risks; resulting in overall improved outcomes, productivity, and cost efficiency. Our solution provides a step-change on how AI solutions are applied in the construction industry from a siloed problem-specific approach to a holistic system-based approach. **Relevance** Our project explores suitability of using Generative AI-enabled systems for the following capabilities: 1. **Data analysis**: analyse large volumes of project data; identifying patterns/anomalies. 2. **Risk prediction**: analyse historical project data; identifying/predicting risks/hazards likely leading to problematic projects. 3. **Real-Time Monitoring**: monitoring ongoing construction projects by integrating data; detecting deviations from the project plan. 4. **Natural Language Processing**: analyse textual project documentation data; extracting valuable insights. 5. **Benchmarking/Comparative Analysis**: compare project data with industry benchmarks/best practices; identifying projects performing below expectations or at higher cost/time overrun risks. 6. **Human-in-the-Loop**: explore inclusion of humans in guiding AI processes and influencing results.

This proposal is a **feasibility study** entailing the following activities: 1. To build a consortium of industry and academic partners to evaluate the suitability of generalisable evaluation methods for Artificial Intelligence (AI) and Machine Learning (ML) approaches. 2. To ensure the commitment of the parties to join a potential submission for the second phase. 3. To propose the development of novel validation and verification approaches for trusted and responsible AI and ML. 4. To produce a technical report outlining the approach to be developed in the second phase. The specific themes that we are targeting are **data pre-processing** and **evaluation**. Our approach is to use the Connected & Automate Vehicle (CAV) market as background and approach to the feasibility study. There is an ever-increasing need to ensure safety and trust for CAVs. It has been proven that it is impossible to ensure safety for CAVs through vehicle-level testing only. Additional challenges exist due to the unbounded conditions of the massively wide operational domain design (OBD); thus, relying exclusively on real-world testing to demonstrate safety across the full spectrum of scenarios that vehicles might encounter in deployment is an impossible proposition. Real-world testing of a vehicle's performance requires access to large test ranges with vast expanses of varied corner and edge case scenarios corresponding to the vehicle's expected OBD. Virtual scenario generation is required for verification and validation (V&V), but the generation of realistic and trustworthy scenarios efficiently remains an open question. A solution is needed that can provide thousands of scenarios and equivalent miles on public roads, which can significantly reduce the development costs and times. **Areas of focus** We intend to conduct a feasibility study to investigate a V&V solution through blended virtual and physical approaches to address the thousands of scenarios and equivalent miles on public roads by providing additional contextual and semantic information of the scenarios, which will significantly reduce the development costs and improve efficiency by ~500 times. **Innovations** Our project proposes the integration of two different approaches, procedural and generative, to the V&V challenges. Our approach will enable the generation of virtual evaluation scenarios to address corner and edge scenarios that are difficult to replicate in real-life. **Relevance** The initial project proposes a solution for developing a service that is generalisable for different Automated Driving Systems (ADS) by providing quick inferences and efficiently generating robust enriched contextual labelled datasets that are enriched with 3D information.

**Introduction** The COVID-19 contingency has highlighted the urgent need for site monitoring systems that enable measuring relative distances, effectively and accurately, among workers and plant equipment to ensure safe working conditions. Government guidelines have been useful, but they are not sufficient. For example, construction sites and manufacturing facilities have been shut down indefinitely due to large numbers of contagions among workers. The construction output fell by more than 40% in April (Office-for-National-Statistics). This underlines the need for accurate and inexpensive monitor systems that contribute to provide safe working conditions while maximizing throughput and productivity. **Areas of focus** Common camera-based site monitoring solutions focus on detecting damage due to crime and environmental hazards, such as intruders and fires. Other more capable systems enable limited object and change detection. Monitoring approaches that make use of Bluetooth devices as mobile phones and wearables have been proposed as well, but they reliability has not been proven yet (e.g. mobile tracking apps). Moreover, they require additional equipment, which increases costs and makes adoption more difficult. **Innovation** This project proposes an innovative approach to develop the datasets necessary to enhance camera-based monitoring systems so as to improve significantly their current capabilities. This project can potentially deliver a qualitative step on the value that monitoring systems provide. For instance, by estimating activity efficiencies, safe distances, and identifying potential contagions. In addition, this project will provide a solution that will not require specialised equipment and will work similarly to current site monitoring systems. **Urgency** This is a timely project, as it will gather very valuable data on currently active sites during the pandemic to gain insights of how monitoring systems can be used to mitigate disruptions due to future sanitary contingencies.

**Introduction** The COVID-19 contingency has highlighted the urgent need for site monitoring systems that enable measuring relative distances, effectively and accurately, among workers and plant equipment to ensure safe working conditions. Government guidelines have been useful, but they are not sufficient. For example, construction sites and manufacturing facilities have been shut down indefinitely due to large numbers of contagions among workers. The construction output fell by more than 40% in April (Office-for-National-Statistics). This underlines the need for accurate and inexpensive monitor systems that contribute to provide safe working conditions while maximizing throughput and productivity. **Areas of focus** Common camera-based site monitoring solutions focus on detecting damage due to crime and environmental hazards, such as intruders and fires. Other more capable systems enable limited object and change detection. Monitoring approaches that make use of Bluetooth devices as mobile phones and wearables have been proposed as well, but they reliability has not been proven yet (e.g. mobile tracking apps). Moreover, they require additional equipment, which increases costs and makes adoption more difficult. **Innovation** This project proposes an innovative approach to develop the datasets necessary to enhance camera-based monitoring systems so as to improve significantly their current capabilities. This project can potentially deliver a qualitative step on the value that monitoring systems provide. For instance, by estimating activity efficiencies, safe distances, and identifying potential contagions. In addition, this project will provide a solution that will not require specialised equipment and will work similarly to current site monitoring systems. **Urgency** This is a timely project, as it will gather very valuable data on currently active sites during the pandemic to gain insights of how monitoring systems can be used to mitigate disruptions due to future sanitary contingencies.