Cost Estimation
AI tools are being used to streamline processes involved with cost planning and quantity take-off in the early phases of construction. Repetitive tasks, such as measuring floor areas, counting building elements, and extracting quantities from plans, can be automated to save time, reduce human error, and improve efficiency.
For example, some AI tools analyse uploaded files- like drawings or BIM models- to automatically identify and measure components. These files often contain sensitive or confidential information, so it is important that any platform used complies with data protection regulations and has appropriate security measures in place. Using secure, trusted platforms and following proper data handling procedures is essential to protect client information. By helping to speed up early-stage tasks like measuring and counting, AI-generated outputs provide a starting point for quantity surveyors. This means they can focus more time on high value tasks like analysis, decision-making, and advising clients.
Though these tools can optimise workflows, there are several limitations. Firstly, the AI model requires high-quality inputs to produce high-quality outputs. Therefore, if the drawings or BIM models are incomplete and/or inaccurate, the AI may extract the wrong quantities or fail to identify important elements, resulting in incorrect cost estimations.
Another issue is that many of these tools are proprietary and not tailored to the specific needs of individual projects and therefore they can lack contextual understanding of the project, such as site constraints or alternative construction methods. As a result, human, professional oversight and expertise is essential. While AI can be beneficial, especially during the initial planning and feasibility stages, it should not be relied on solely for making decisions. Final estimates must always be reviewed and validated by qualified professionals to ensure that they are accurate and reliable considering the context of the project.
Project Scheduling
Within construction, AI systems are also used to help predict project timelines, optimise processes and reduce delays. These systems use predictive models to analyse a variety of historic and real time data and identify patterns, predict issues and make data-driven decisions to help mitigate risks.
Depending on the tool and how it is used, the outputs can vary. For example, some systems may provide suggested timelines, risk alerts, resource allocation plans, or scenario comparisons to help project managers make more informed decisions. This information is typically visualised through dashboards, charts, or Gantt-style timelines, helping project managers quickly assess possible outcomes and impacts.
By analysing such a large volume of data at scale, AI systems can identify risks that may be overlooked when performed manually. This provides project managers with a clearer overview of risks, allowing them to make proactive decisions earlier—potentially saving both time and cost.
However, AI systems often lack knowledge and understanding of the project-specific context, such as site conditions or stakeholder expectations. Therefore, it is essential that the system is trained on high-quality, relevant and detailed data. If the system is trained on incomplete or inaccurate information, then it could provide misleading outputs or poor recommendations, potentially resulting in cost overruns or delays and misinformed decisions.
To reduce these risks, AI tools should be used to support, not replace, human decision-making. While the output can provide helpful insights, the project manager must still interpret this information in context and make final decisions based on professional judgement. At its best, AI-powered scheduling tools act as an assistant, helping to automate repetitive tasks such as sequencing and resource allocation. However, project managers remain responsible for ensuring that schedules remain realistic, flexible, and fit for purpose.
Principal author: Aaliyah Pollock