Artificial intelligence may be a trendy tech buzzword but it's been used in transportation safety for a long time.
Joining Jason and Matt on this week's 10-44 is Netradyne Director of Performance Marketing Austin Schmidt who says AI's role in trucking can be more than just dash cam footage that keeps fleets out of court; it can be an enabler of efficiency too.
Contents of this video00:00 Safety and efficiency in trucking through artificial intelligence
02:45 Having good artificial intelligence
05:53 Collecting driver data beyond visuals
06:47 Artificial intelligence and driver privacy
08:47 Using advanced fleet safety technology to influence direct operational costs
This week's 10-44 is brought to you by Chevron Delo 600 ADF ultra-low ash diesel engine oil. It's time to kick some ash.
Artificial intelligence isn't just a techy buzzword. It's a key enabler of trucking safety and efficiency.
Hey everybody and welcome back to the 10-44, a weekly web episode from the editors right here at CCJ. I'm Jason Cannon and my co-host on the other side is Matt Cole.
You hear words like artificial intelligence and machine learning thrown around a lot. Often it's about how Facebook and Amazon know the next thing you want to buy even before you do. AI's played a role in trucking safety for a long time. Take dash cams for example. Netradyne Director of Performance Marketing, Austin Schmidt, who joins us on the 10-44 this week, said a dash camera with good AI capabilities makes the camera itself a better tool, capable of doing more than simply just keeping the carrier from losing a lawsuit.
Originally dash cams were pretty much just for exoneration. It's this crash for cash schemes. They're just basically showing what happened, basically getting the fleet off the hook when they aren't at fault in the case of an accident. This is inherently reactive.
As the technology itself started to really advance, we got things like critical events, hard breaking events, hard acceleration events, maybe a hard corner. This had some utility for fleets as they could now identify their most dangerous drivers, what shows formally a complete blind spot to them. It started to get a little bit more reactive. The problem here is a hard break doesn't always explain exactly what happens, it doesn't understand the context of the event. As you can imagine, if you are driving a heavy duty truck and a passenger vehicle merges onto the interstate in front of you and you have to hit the hard brakes, if you get dinged in that in your quarterly safety meeting, you're not going to be thrilled with that because that's actually great driving and defensive driving.
Now the background, we start getting into AI started making it to market. When AI started to make it to market, we built on this... Speaking we as the industry, not Netradyne here, we started to build on new events like distracted driving, following distance violations, that sort of thing. We were scoring it in the same manner. When you think about it from the example we just gave, that actually accentuates parts of the context problem. It's going to sound somewhat humorous, but did the driver grab their phone or did they grab a can of iced coffee? This sort of inaccuracy is actually commonplace with some systems in the market, and you can imagine that's really going to upset your great drivers.
Austin stresses that simply having AI isn't good enough, it's having good AI. As a fleet, how do I know the good from the bad?
When I think about it, so what really is AI wanting to do? It's wanting to... This is speaking very loosely, AI is trying to automate things that humans would typically do with this our natural way of thinking and problem solving. When the AI says that the driver was on their phone, what percentage of those alerts was the driver actually on their phone? Events like that, I would really think about it as this accuracy when the camera says that something happened, is that actually what happens when a human reviews it? I think that's the answer. One, I would talk to customers that are currently using the product. Many of them will have pretty strong opinions, whether positive or negative on whether their experience has been that the AI is accurate. Two, if you want to go a little bit deeper, try the product, review the videos and the events for yourself and just put them to the test.
All artificial intelligence are not equal. I think sometimes it can be easy, just every fleet camera out there right now, because this became a buzzword, is saying we have artificial intelligence. It's not necessarily a box to check on your evaluation checklist because it really is that quality, and basically on quality, I would think about it just like anything that exists in the world, it really is the quality of the people that build it and the quality of people that optimize it. Great engineers develop great AI and great machine learning algorithms to optimize that AI, which leads to great results. Which is why I think you've just got to talk to customers who are using the product, understand their experience with that AI accuracy, and just test it for yourself because that's really how you're going to know if it's going to work for what your business needs.
AI has its place at the fleet level beyond driver camera systems too. Austin tells us where after a word from 10-44 sponsor, Chevron Lubricants.
Protecting your diesel engine and its aftertreatment system has traditionally been a double-edged sword. The same engine oil that is so essential to protecting your engine's internal parts is also responsible for 90% of the ash that is clogging up your DPF and upping your fuel and maintenance costs. Outdated industry thinking still sees a trade-off between engine and emission system protection and Chevron was tired of it, so they spent a decade of R&D developing a no compromise formulation. Chevron lubricants developed a new ultra low ash diesel engine oil that is specifically designed to combat DPF ash clogging. Delo 600 ADF with Omnimax technology cuts sulfate ash by a whopping 60%, which reduces the rate of DPF clogging and extends DPF service life by two and a half times. Just think what you can do with all the MPGs you're going to add from cutting your number of re-gens.
Delo 600 ADF isn't just about after treatment. It provides complete protection, extending drain intervals by preventing oil breakdown. Before, you had to choose between protecting your engine or your after treatment system, and now you don't. 600 ADF from Delo with Omnimax technology, it's time to kick some ash.
When you connect to the engine control module, you're able to collect a lot of data about your drivers. Not just visual data, everything from the speed to diagnostic information from the vehicle and the engine. There are a lot of ways that you can analyze trends in the data because when you're trying to optimize a cost structure of a business, a lot of it is trying to put probabilities in your favor. Let's say that an engine had a higher than normal number of hard breaking and hard acceleration events over the last quarter. It's very possible that that wear and tear is going to lead to that engine needing to be repaired more frequently. There are a lot of use cases for artificial intelligence in terms of making a fleet more efficient, rather than just the visual part. I definitely see a lot of use cases moving forward. It's going to be a pretty exciting time in fleet tech.
One of the best parts of AI is how efficient it can make reviewing incidents by basically filtering it. Fleet managers don't have to watch hours of video just to see that their driver's harsh braking incident was an evasive maneuver and in no way the driver's fault. Rather, they can hone in on those coachable moments while also giving the driver a greater degree of privacy.
Let's just say that I am driving, and Jason, you are the fleet manager. It's a one truck operation on this. You believe that distracted driving is a potential significant risk for your business, and you want to understand if Austin, your newly hired driver is texting and driving. Without AI, a human's going to have to go and review all of that video to look for texting and driving. The AI is basically taking that human recognition of an event and then surfacing that for you. It's eliminating the ability for humans having to go and manually review your review video footage for specific event that matters to your business. Think about it really as efficiency.
To having really accurate AI, there's also actually privacy benefits from a driver's perspective. When you have accurate AI that you can rely on that's going to say, "Hey, this is accurate to what happened," you don't need a fleet manager going through and just watching all of this footage from the cab. Me as a driver in this situation for Jason's fleet, that would probably make me a little bit more comfortable, provided that I have confidence that that artificial intelligence is accurately scoring my driving. From that perspective, I think that AI is about efficiency and also from me as the driver in the scenario, I know that you, Jason, don't have to sit there and just look at all of my video either, which just, to me at least, makes me actually a little more comfortable in the cab.
As AI technology evolves and gets more reliable, Austin says it becomes more applicable outside the realm of safety,
Transitioning your safety department from a cost center to a cost saver. What we mean by that, is by using advanced fleet safety technology, there's an ability to influence direct operational costs for your business. It's not just, "Hey, it might reduce an accident maybe quarters or years down the road, which would save me a lot of money." Could be a high probability, could be a low probability depending on your business, but think things like fuel. There are multiple studies that show that aggressive driving, so hard braking, speeding, that sort of thing reduces fuel economy by 15 to 30%. I would ask everybody to do, okay, look at what you spend on fuel, just put into a quick Excel sheet, put in 3%, 5% fuel efficiency. What kind of impact would that have on your bottom line, and how does that monetary value compare to the cost of the safety technology?
Basically, when you think about an advanced solution that can accurately not just do the more basic things like speeding, but accurately say, "Hey, please leave a little bit more space in between that next vehicle, which reduces the chances you're going to have to hit on the hard brakes." Live in-cab coaching can really help fleets use a lot less fuel, which can help really improve operating margins in the short term.
That's not just the only thing, it's probably the most concise and clear cut example, but things like driver retention, solutions that can reinforce positive driving and be accurate on the scoring are shown to be able to actually decrease driver retention. We all know that replacing a great driver is very expensive and almost impossible. I think you look at the impact that this type of technology can have on fuel, driver retention, and other operational costs as well, then I think that that's how this technology is evolving. The safety impact is massive, but the operational cost impact if you choose the right solution is incredibly underrated.
That's it for this week's 10-44. You can read more on CCJdigital.com, and as always, you can find the 10-44 each week on CCJ's YouTube channel. If you've got questions, comments, criticisms or feedback, please hit us up at email@example.com or give us a call at 404-491-1380. Until next week, everybody, stay safe.