Like all firms, it is important for Engineering News Record (ENR) to know if unsafe conditions are being shown in the imagery it releases to the public. For their annual photo competition, the ENR team used our machine learning to review all 763 images submitted for consideration. In this article published by ENR, you can see our ML engine, nicknamed “Vinnie”, found 97 people not wearing gloves, 31 without hardhats, 89 without safety colors and 11 without eyewear. We helped screen the winning pictures and identified two that we recommended be reviewed further.
Vinnie detects missing safety colors; image courtesy of ENR
ENR is not alone in using machine learning to impact their business. Machine learning is talked about seemingly everywhere -- from McKinsey studies on the potential of machine learning for businesses to the construction industry tech report from JB Knowledge.
Not surprisingly, larger companies are taking the lead. Bechtel has staffed up a machine learning center of excellence and already has saved millions. Suffolk construction has staffed up highly capable machine learning teams with MIT PhDs. But not every company can afford to build a “center of excellence” like Bechtel, or create a ML team with PhDs like Suffolk.
This reminds me of another technology shift that impacted the AEC industry - the introduction of mobile applications on iPads and cloud computing. Back in the “old days”, I worked with executives and field personnel to bring mobility and cloud to construction at Vela Systems (now Autodesk BIM 360 Field). This technology wound up being more than just buzz words in the late 2000’s, the systems we built helped change the way quality, safety and commissioning are done today. Most teams wouldn’t imagine going back to doing punchlists in Excel or giving back their iPads and bringing back drawing tables and racks on every floor of the job.
We see a similar opportunity now with machine learning. The key is that the technology must tie to real business value. For this reason, our vision is tied to three key aspects of construction delivery (whatever the type): safety, productivity and quality.
Wondering how ML can help your business? Work with us in a “machine learning guided tour” where we can help explore how we can help change your business. Take glove compliance for example. Many companies have 100% glove compliance initiatives but struggle to know if teams are in compliance. Just like we helped ENR, we can help you analyze all of your photos from all of your projects, creating what one safety manager called “an extra set of eyes that doesn’t sleep” to help reduce risk. The result: fewer hand lacerations and related costly injuries.
Sound expensive? Sound technologically challenging? It’s not either of those things. Keep your photos and/or videos in Procore, Autodesk, Egnyte, or - now in Beta testing for early 2018 - Oxblue cameras. Get started in less than 90 seconds through one of our pre-built adapters to these systems.
Did you wait to try the iPad at your construction sites? Did you wait to try the cloud? Don’t wait to try ML. You’ve already started down the road of technology. Be sure to continue to maximize the value of all of the data you’ve collected (and are continuing to collect). ENR is using it to help reduce risk - shouldn’t you?
