{"id":10506,"date":"2021-04-01T13:49:43","date_gmt":"2021-04-01T11:49:43","guid":{"rendered":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/?p=10506"},"modified":"2021-04-01T13:49:43","modified_gmt":"2021-04-01T11:49:43","slug":"diagnostics","status":"publish","type":"post","link":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/2021\/04\/01\/diagnostics\/","title":{"rendered":"Making Diagnostics Smarter"},"content":{"rendered":"

Photo:\u00a0Rush Enterprises<\/i><\/span><\/p>\n

Rush Enterprises says it\u2019s starting to gather data from multiple components and use it to understand how and when to service vehicles.\u00a0<\/b><\/span><\/p>\n

New, powerful technological forces in play are using data to transform truck maintenance and diagnostics.<\/span><\/p>\n

The two main drivers are artificial intelligence (AI) and machine learning. When combined with vehicle telematics systems and the Internet of Things, these two, relatively new, computing capabilities will soon allow trucks and even components to continually monitor and seek ways to improve their operational health and alert maintenance professionals when a critical failure is imminent.<\/span><\/p>\n

\u201cFor example, a machine learning model can track the trend of battery cranking voltage over a period of time (60-90 days) in relation to other relevant data points such as ambient temperature,\u201d explains Braden Pastalaniec, vice president of sales for transportation and logistics for Uptake. \u201cWhen a decreasing trend is observed beyond a certain threshold, the fleet operator is alerted to take immediate action on that particular asset.\u201d Uptake provides an artificial intelligence platform that turns the untapped data generated by equipment into actionable insights.<\/span><\/p>\n

The possibilities of these powerful tools in fleet maintenance operations are obvious: A water pump, for example, that can alert a fleet that it is failing is an incredibly powerful asset. But a water pump that can predict its own imminent demise based on past performance and operational data and can help a fleet manager schedule maintenance two weeks before the failure is likely to occur is a next-level empowerment for fleet maintenance professionals.<\/span><\/p>\n

\u201cIf you examine the history of heavy-duty truck maintenance, you\u2019ll notice that diagnostics have become smarter each year,\u201d says Scott Bolt, vice president, product management, Noregon. \u201cWe\u2019ve gone from flash codes to electronic component-based software, to comprehensive aftermarket solutions that simultaneously diagnose all components. Even within that aftermarket software we see improvements each year with the development of technology like guided diagnostics and repair assistance.\u201d<\/span><\/p>\n

Building on those trends, Bolt says, is an \u201cunprecedented\u201d growth of smarter diagnostics driven by telematics and the internet of things. Whether through telematics services, electronic logging devices, or remote diagnostic hardware such as Noregon\u2019s ND2 device, he notes, fleets no longer have to wait for an in-shop visit to collect vehicle data. It is now captured continuously while the truck is in operation.<\/span><\/p>\n

\u201cA major advancement we\u2019ve recently seen in smart diagnostics is the emergence of predictive fault technology,\u201d Bolt says. \u201cFor example, in our remote diagnostic application, TripVision Uptime, we now present users with an alert when a fault has the potential to escalate to a more serious issue, with a percentage likelihood of escalation and a time and mileage estimate for when the escalation would occur. This information helps fleets start to build maintenance programs that move away from interval-based maintenance and develop a condition-based maintenance strategy.\u201d<\/span><\/p>\n

\n

\"Although<\/span>
Although technicians are already using high-tech diagnostic tools like Noregon\u2019s JPro in the shop, artificial intelligence and machine learning are on the cusp of transforming fleet maintenance.<\/span><\/p>\n

Photo:\u00a0Noregon<\/strong><\/span><\/p>\n<\/figcaption><\/figure>\n

Fleets and dealerships can already pull fault codes off a vehicle with telematics, notes Victor Cummings, vice president of service operations, Rush Enterprises. \u201cBut the challenge now is to make that information more accessible. We need to be able to take it and marry it with a vehicle\u2019s repair history so that, for the first time ever, we\u2019re going to be able to see and understand why the vehicle has an issue, what needs to be done to correct it, and then seamlessly add in everything needed to make that happen \u2013 from scheduling a repair, to assigning the correct technician to the job, to making sure the parts are available at that time, and then building an accurate repair estimate so the fleet knows all of these things before the truck has even been taken out of service.<\/span><\/p>\n

\u201cWe are really in the exploratory stages of how we\u2019re going to connect all the different systems and components on a truck.\u201d<\/span><\/p>\n

Transformative technology<\/span><\/h2>\n

Once that exploratory stage is complete, adds Cindy Hunter, technology sales director for Rush Enterprises, the next step will be a seamless, fully integrated, cross-platform means of collecting, transmitting and presenting all that data in a user-friendly, actionable format.<\/span><\/p>\n

\u201cThat is the next level that we are working towards today,\u201d she says. \u201cWe are starting to gather information off of multiple components \u2013 everything from tires, to wheel ends, to brake linings, to wiring, to fuel filters. And we\u2019re starting to use this data now to understand how and when to service these vehicles. That process has already begun, and we are continually stacking new layers of sensors, technology and data mining on top of the systems we already have in place.\u201d<\/span><\/p>\n

Technologies such as self-diagnosis and machine learning will lead to maintenance practices that are more cost-effective and improve the life of the truck and its components, Bolt adds. \u201cUsing these technologies to develop prognostics and implement advanced maintenance practices like condition-based maintenance empowers fleets to maintain a truck based on how it is used, rather than a time or mileage interval. Instead of \u2018this truck drove 20,000 miles since its last oil change,\u2019 fleets can base the decision on where the truck operated, how many cold starts occurred, and use sensor data to help dictate the optimal time for the oil change.\u201d<\/span><\/p>\n

Bolt says the recent explosion in machine learning is made possible by the vast amounts of data we\u2019re now able to capture from vehicles. More than ever before, trucks are online and transmitting data, meaning fleets don\u2019t have to bring the truck in to collect that information. By capturing data from the truck while in various operating conditions, fleets can develop a better understanding of how their trucks perform and begin to forecast future maintenance needs.<\/span><\/p>\n

Uptake\u2019s Pastalaniec says its customers (which range from small and medium-size fleets up to giant fleets such as Frito-Lay and PepsiCo) are already seeing a 20% reduction in certain roadside breakdowns, an 8% improvement in technician efficiency, and an overall reduction in maintenance costs by 7-10%.<\/span><\/p>\n

And that\u2019s just the beginning, he says. \u201cA thing to keep in mind is that machine learning improves over time with the ingestion of more data and feedback on predictive outcomes, so we\u2019re just beginning to see the promise of industrial AI.\u201d<\/span><\/p>\n

In time, Pastalaniec predicts, \u201cwe\u2019ll see a maintenance process that seamlessly begins when AI determines the time is right for a maintenance event based on its financial impact or other mission-critical benchmarks \u2013 whether it\u2019s a failure symptom picked up from the on-board computer, a scheduled preventive maintenance task, or a component running into its last 5% of useful life. From that point, the asset is scheduled for service at the optimal shop, where parts begin to be pulled, and the technician is given full diagnostic information based on the most probable failure. When the asset arrives, the technician is virtually assisted through the repair and all work digitally recorded, and all paperwork is automatically filed.\u201d<\/span><\/p>\n

Beyond the shop<\/span><\/h2>\n

Building upon those capabilities will give fleets opportunities to collect and analyze aggregated data across an entire fleet of vehicles, rather than on a truck-by-truck basis. By noticing trends in similar makes and models of trucks, fleets can improve purchasing decisions and better determine overall maintenance needs.<\/span><\/p>\n

Fleets can determine, for example, if a particular make or model of truck, with a particular powertrain package, performs better on certain routes or geographic areas of the country than other makes, models or powertrain combinations. They\u2019ll be able to determine which drivers are not just good performers on fuel economy, but also which ones drive trucks in a way that minimizes wear and tear and holds maintenance costs down.<\/span><\/p>\n

And that\u2019s not all. Soon fleets will know, for example, that a certain series of water coolant tanks installed during vehicle assembly during a specific time frame are prone to cracking and leaking at a consistent mileage point during the life of the vehicle. They\u2019ll know which brake linings perform better for their applications, and which coolant blends perform the best for them in both cold and hot climates.<\/span><\/p>\n

There will be no need for guesswork or even benchmarking, because the trucks will be providing data to dealers, OEMs and private fleets. And any problem or failure event will be duly noted and recorded and logged as a data point for both an individual truck, other similar makes and models of trucks, and all trucks throughout a fleet. That data will be shared and used by the fleet itself, as well as by dealerships and OEMs, to constantly improve vehicle performance and maintenance procedures from the moment a truck rolls off the assembly line to the day it\u2019s finally placed out of service and hauled off to the scrap yard.<\/span><\/p>\n

\u201cIn many ways, this seems like magic,\u201d says Chris Orban, vice president, data science, Trimble. \u201cFor the first time ever, we are on the cusp of truly understanding how and why vehicles work and perform during their service lives. We really don\u2019t have a true understanding of how environments or being driven in a certain way affects a vehicle over the course of its life right now. We are now starting to add layer upon layer of data to that picture, and the possibilities are endless.<\/span><\/p>\n

\u201cMaybe, for example, that truck in your fleet famous for being a \u2018lemon\u2019 just needs to be driven by a better driver in a less-stress operating environment to perform at its highest possible potential. Very soon, we\u2019ll be able to determine things about fleet operations down to that level.\u201d<\/span><\/p>\n

Smarter diagnostic technology will help change the entire industry \u2013 indeed, is already doing so, Bolt says. \u201cMachine learning, self-diagnosis, self-healing, and other advanced technologies driven by IoT are major steppingstones for the industry to reach various levels of autonomy,\u201d he notes. \u201cRather than being restricted by hours of service regulations, trucks can eventually reach a point where they operate nearly 24 hours per day, which greatly improves the supply chain. The increase in a truck\u2019s operational hours will lead to different maintenance needs than we see today, with an increasing reliance on the truck\u2019s ability to self-diagnose.\u201d<\/span><\/p>\n

Rush recently began working with some of its fleet customers to start using all of this new data coming in off of vehicles outside of the maintenance shop.<\/span><\/p>\n

\u201cBased on requests from those customers, we\u2019ve been pulling data and helping them use it to improve performance in a variety of new ways,\u201d Cummings says. \u201c They want to compare fuel efficiency performance on different routes and help with asset utilization. They\u2019re interested in seeing which of their trucks perform better in certain applications with fewer fault codes. So, the power of this new technology is already being put to work in new ways by some forward-looking fleets now.\u201d<\/span><\/p>\n

Recalibrating mindsets<\/span><\/h2>\n

With all the potential benefits, why aren\u2019t these systems being adopted into fleet operations at warp speed?<\/span><\/p>\n

For starters, as Hunter notes, the technology is still very new and being applied unevenly. Major strides need to be made in crunching the massive amounts of data coming into fleets, dealers and OEMs and putting them in an easily accessible and usage format.<\/span><\/p>\n

But perhaps a larger obstacle is changing the mindsets in many fleets, say both Hunter and Orban.<\/span><\/p>\n

\u201cThere\u2019s no question fleet managers instantly grasp and understand the potential for these new technologies,\u201d Orban says. \u201cThey understand just how powerful and transformative these changes are going to be. But to bring a truck in with no obvious problem for service \u2013 based on an algorithm telling you a problem is \u2018probably\u2019 going to occur in the near future? That\u2019s still a hard leap for many fleet managers to make. Because they\u2019ve been trained and conditioned by years of pressure to deliver uptime to keep trucks in service as long as possible \u2013 right up until the last possible second it can still do its job. There are still people out there who have a hard time accepting the predictive model did its job, because nothing bad ever happened to the truck \u2013 which is exactly how the predictive model is supposed to work.\u201d<\/span><\/p>\n

One other problem Orban points to is that the predictive models and the data coming off of trucks is already so detailed and looks so far into the future, that many fleet managers and technicians have a hard time believing they can be that accurate.<\/span><\/p>\n

\u201cSo, one key issue we\u2019re going to have to address is putting this information in a way so that humans can better understand the \u2018how\u2019 and \u2018why\u2019 of what they\u2019re being told, and make the necessary decisions with greater confidence that they\u2019re actually anticipating and counteracting real, pending maintenance issues.\u201d<\/span><\/p>\n

Hunter concedes that there is a lag in terms of understanding the potential of new diagnostic technologies and actually buying into them. But, she says, the data itself is already proving that these new concepts are valid.<\/span><\/p>\n

\u201cWe had a conversation with Rush\u2019s largest fleet customer about this just the other day,\u201d Hunter says. \u201cAfter reviewing the data from last year, they told us proactive, predictive maintenance saved them between 20 and 30 diesel engines they would have otherwise lost due to maintenance failures.\u201d<\/span><\/p>\n

Pastalaniec says Uptake sees 2021 as \u201ca critical inflection point\u201d for this type of software, \u201cwith some of the early adopters recognizing the value and more OEMs and fleet operators getting onboard.\u201d\u00a0\u00a0\u00a0<\/span><\/p>\n

In short, diagnostics and preventive maintenance are on the cusp of a powerful new information age that will eventually spill over into every aspect of running a successful trucking company.<\/span><\/p>\n

\n

Smart Diagnostics Defined<\/span><\/h2>\n

According to IBM, artificial intelligence \u201crefers to any human-like intelligence exhibited by a computer, robot, or other machine. In popular usage, artificial intelligence refers to the ability of a computer or machine to mimic the capabilities of the human mind \u2014 learning from examples and experience, recognizing objects, understanding and responding to language, making decisions, solving problems \u2014 and combining these and other capabilities to perform functions a human might perform, such as greeting a hotel guest or driving a car.\u201d<\/span><\/p>\n

Machine learning an application of AI. It\u2019s the use of computer algorithms to identify trends \u2013 algorithms that improve automatically through experience as the system \u201clearns.\u201d As it relates to truck maintenance, it\u2019s the ability for algorithms to identify diagnostic data trends, in real time, that previously would have led to equipment failure and downtime.<\/span><\/p>\n

Machine learning is an extension of another data-mining technology, predictive analytics; both make predictions about the future based on data. But the \u201clearning\u201d part means a machine learning model can fine-tune the parameters based on what it\u2019s learning from the data.<\/span><\/p>\n<\/div>\n

\u00a0This article originally appeared in the March print edition of Heavy Duty Trucking.<\/em><\/span><\/p>\n

 <\/p>\n

by Jack Roberts<\/a><\/p>\n

\n
\n

Source: https:\/\/www.truckinginfo.com<\/a><\/span><\/p>\n<\/div>\n<\/div>\n

\n
\n
\n
\n
\n

CUT COTS OF THE FLEET WITH OUR AUDIT PROGRAM<\/strong><\/a><\/h3>\n

\"\"<\/a><\/p>\n

The audit is a key tool to know the overall status and provide the analysis, the assessment, the advice, the suggestions and the actions to take in order to cut costs and increase the efficiency and efficacy of the fleet. We propose the following fleet management audit.<\/p>\n

FLEET MANAGEMENT AUDIT<\/strong><\/a><\/h3>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n","protected":false},"excerpt":{"rendered":"

Photo:\u00a0Rush Enterprises Rush Enterprises says it\u2019s starting to gather data from multiple components and use it to understand how and when to service vehicles.\u00a0 New, powerful technological forces in play are using data to transform truck maintenance and diagnostics. The two main drivers are artificial intelligence (AI) and machine learning. When combined with vehicle telematics…<\/p>\n","protected":false},"author":3,"featured_media":10507,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[18],"tags":[181,53],"_links":{"self":[{"href":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/wp-json\/wp\/v2\/posts\/10506"}],"collection":[{"href":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/wp-json\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/wp-json\/wp\/v2\/comments?post=10506"}],"version-history":[{"count":1,"href":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/wp-json\/wp\/v2\/posts\/10506\/revisions"}],"predecessor-version":[{"id":10508,"href":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/wp-json\/wp\/v2\/posts\/10506\/revisions\/10508"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/wp-json\/wp\/v2\/media\/10507"}],"wp:attachment":[{"href":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/wp-json\/wp\/v2\/media?parent=10506"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/wp-json\/wp\/v2\/categories?post=10506"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/wp-json\/wp\/v2\/tags?post=10506"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}