The modern rise of data science has seen the use of machine learning (ML) and artificial intelligence (AI) to provide various industries with the power of automation and predictive analysis. Industry 4.0 or the Fourth Industrial Revolution has moved forward in parallel, providing a digital transformation framework enabling manufacturing, production, and engineering companies to aim towards greater levels of efficiency through integration of ML and AI. With the lens on civil engineering specifically, Boulevard has developed an open-source ML-based sample solution showcasing readiness support to aid organizations further automate their processes and development of chemical compounds.
Concrete is one of the most used materials in modern civil engineering efforts as individual projects require significant amounts of concrete with various strength requirements. Evaluating concrete strength is a time-consuming and largely manual process. Typically, concrete strength can only be evaluated via samples after a batch has been made resulting in a continuous waste in both time and resources. Our team at Boulevard Consulting has developed a ML-based solution to reduce waste and increase concrete-making efficiency.
The custom ML application, named Compressive Strength Predictor, provides civil engineering companies with a digital sandbox for evaluating concrete strength. Using this application, users can fully customize ingredients for a concrete mixture and dynamically receive an accurate estimate of concrete compressive strength in Megapascals (MPa) in a matter of seconds rather than waiting days, weeks, and even months for manually evaluating samples. While manual strength testing should still be part of every company’s processes as means of final validation, the Compressive Strength Predictor provides an efficient way to digitally test samples without requiring additional resources to be used. This ultimately provides users with the means to navigate the infinite combinations of developing concrete mixtures and quickly identify what a likely end-state composition would look like for the users’ given purpose.
While the capability to automate compressive strength evaluations of individual mixtures is useful, our application has proceeded to move one step further. In addition to using our ML model to predict the strength of individual mixtures, our application has integrated this model within an optimization framework allowing users to specify target strength requirements and ultimately have the application generate a concrete mixture for them, with specifications on amounts of each component needed. The ability to autonomously generate optimized mixtures based off user-defined requirements will allow our application to fully support projects end-to-end. Whether it is using our applications during the planning phase to develop expected resource needs for the entire scope of a build, or a quick turn strength evaluation of a single proposed mixture, we aim to provide automation and added transparency through every stage of a development effort.
We at the Boulevard Consulting Group pride ourselves on providing not only the technical expertise to utilize the latest in ML and analytics techniques, but also the strategy and operations experience to craft solutions customized and optimized for each client. To learn more about the AI/ML principles discussed within this article, visit Boulevard Consulting.
See below to see the Compressive Strength Predictor in action.
コメント