How machine learning can analyze your job data, prevent risk

Machine learning is a subset of artificial intelligence (AI). It uses algorithms that tell computers how to interpret data without having to be explicitly programmed to solve a specific problem. 

In construction developments, Newmetrix  (formerly Smartvid.io) is developing image recognition software to distinguish features on jobsite photographs that represent safety risks. Autodesk has introduced Pype, which uses SmartPlans to identify and organize submittals. But, when it comes to risk, the most tantalizing development in the past few years came from the Autodesk world. Construction IQ (formerly BIM 360 Project IQ) uses AI to provide insight into risk on projects. Described by an Autodesk Digital Builder article, Construction IQ, part of the Autodesk Construction Cloud (ACC), has built-in computing technology that gives project teams insight into predicting risk, decreasing it and managing it. The program studies jobsites for issues that fit into the four primary causes of jobsite accidents: falls, struck-by, caught in/between, and electrocution. It reports out high-risk safety issues so they can be addressed before accidents happen.

Construction IQ also tracks project quality. It will flag issues early in the job that it interprets as having risk, for instance inspection problems, anticipated rework, design or documentation issues, code compliance, errors and omissions. The root cause of requests for information (RFIs) is often in design or documentation. Construction IQ can identify and prioritize these issues faster than a manual process using the Document Management module to initiate organized, multi-step approval workflows.

With Construction IQ, risk data is collected using Autodesk Build. Project data is scanned using machine learning and artificial intelligence (AI) in the Design Risk Management feature of Construction IQ. The process quickly reports out the highest risk issues on a dashboard, as shown in the following Autodesk video.

RFIs can be prioritized by Construction IQ using a risk scale. Since RFIs have been found to be the main contributor to cost overruns, schedule delays and litigation, addressing this risk is critical to keeping a project on track.

According to Autodesk FAQs, cost impact and quality risk were developed as separate models in Construction IQ. Issues that are identified may have a high cost impact with a low risk level, low cost with high risk, and so on. The suggestion is that issues with high cost plus high risk level usually receive top priority.

According to Autodesk Knowledge Network as of January 2022, for access to Construction IQ, end users need to have Autodesk Build accounts with these criteria:

• Managing quality and safety analytics – use Project Management.
• Design review analytics – use design issues in Document Management.
• Project types – Construction IQ will currently analyze commercial, residential, institutional, healthcare, and transportation building infrastructure projects.
• Accounts in English – Construction IQ machine learning algorithms currently work only with data captured in the English language.
• Accounts in the US data center – for users of Classic BIM 360 Field.
• Accounts in US or Europe-Middle East-Africa data centers – for users of Autodesk Build and BIM 360 Build.

---

If you need a partner to help you investigate or implement machine learning, contact Applied Software today. The experts of Applied will help you champion the solution that is right for your company.