The construction industry is a significant contributor to global carbon dioxide (CO2) emissions, with estimates suggesting that it is responsible for approximately 43% of global energy-related CO2 emissions (World Green Building Council, 2022). In order to combat climate change, it is crucial that we accurately track and reduce these emissions.
Calculating the CO2 footprint of a construction project can be a labor-intensive process, requiring manual data entry and analysis. This can be prone to errors and inconsistencies, and it is difficult to track and compare emissions across multiple projects. This challenge is general in the sector and is exacerbated by the poor level of digitization, which translates into two very important technical challenges: (a) the fragmentation of data sources produced by the use of multiple independent digital tools; and (b) the inability to analyze these data in a cross-cutting and consistent manner due to the lack of interoperability between these digital tools. In addition, the data is often unstructured and there is a very significant heterogeneity that prevents cross-cutting analysis of the information.
One way to optimize this is by using artificial intelligence (AI) techniques to extract and analyze data about the materials used in construction projects, which can be used to calculate the CO2 footprint of each project. This can be done by scanning documents such as invoices, contracts, and drawings, and using algorithms to extract relevant information about the materials used.
The use of AI in this context has several benefits. Firstly, it is much faster and more accurate than manual data entry, as AI algorithms can process large amounts of data in a short amount of time and with a high degree of accuracy. Secondly, it allows for real-time tracking of emissions, as AI algorithms can continuously process new data as it becomes available. This means that the CO2 footprint of a project can be tracked and updated throughout the construction process, rather than waiting until the end of the project to calculate the emissions.
In addition to tracking emissions, AI can also be used to optimize the use of materials in construction projects. For example, algorithms can be used to identify opportunities to substitute materials with lower carbon emissions or to find more efficient ways of using materials. This can help to further reduce the CO2 footprint of a project, and can also result in cost savings for the project.
Practical automation use case to better control CO2 emissions
We recently worked with a company that was seeking an innovative approach to better control the CO2 footprint of materials purchased by the organization during the project phase, with the goal of understanding the real footprint of each of the company's projects produced by the materials used.
It is possible to associate an estimated carbon footprint with almost any type of material (m3 of concrete, m3 of sand, etc.). However, the information on the materials used is not collected in a format that facilitates access to the information. This information is usually contained in invoices issued by suppliers, typically in PDF format, and each supplier uses their own unique invoice model which makes it difficult to automate the process of extracting information from these invoices and creates challenges in developing a generic system that can extract the information from invoices in previously unseen formats (such as from a new supplier).
At Beawre, we found an automation solution to this problem. We automate the extraction of information from heterogeneous-format invoices using AI techniques, combining optical character recognition (OCR) capabilities with machine learning algorithms to analyze the structure and layout of supplier invoices and pattern recognition techniques. Fields and lists of materials can be analyzed and extracted, and the carbon footprint calculation for each project automated based on invoice data to create a continuous control mechanism during the project phase.
We train the AI models used for automated invoice processing to learn how to extract relevant information for calculating the carbon footprint of all materials in the context of construction projects and to extract significant data for this purpose from the invoice, regardless of its format. Our technology can connect to public databases, such as those published by the Catalan Institute of Construction Technology Foundation - ITeC, and private databases like SIMAPRO, GABI, and UMBERTO, to calculate the carbon footprint.
Conclusions
The use of AI techniques to extract and analyze information about the use of materials in construction projects can be a powerful tool for accurately tracking and reducing CO2 emissions. By using these techniques, it is possible to not only calculate the CO2 footprint of a project but also to optimize the use of materials and continuously track and reduce emissions throughout the construction process.
References:
World Green Building Council. (2022). WorldGBC Advancing Net Zero Status Report 2022. Retrieved from https://www.worldgbc.org/resources