The vision is a world with seamless integration of technology. Everything is intelligent. Everything is wireless. Everything works. Equipment operators are happy and productive, and business owners are steeping in profit.

The reality is often a world in which levels of technology are uneven. Nothing talks to anything else. The systems have more moods than a rabid Chihuahua. Operators fear the stuff, and owners rue every nickel they tossed into the technology sinkhole.

What went wrong?

"Nothing is easy in technology," said Pam Nelson, CEO of CCG Systems. CCG makes FASTER software for the management of fleet-maintenance operations. Its 17 modules track labor, work orders, inventory, and fuel. Nelson has edited a monthly fleet-management publication for 14 years and is often called upon to give presentations on surviving technology to organizations such as the Equipment Maintenance Council. "Nothing plugs and plays."

Nothing? It seems that way. Even if the installed application runs fine, it often has a negative impact on other programs already in place.

That's not to say there isn't some phenomenal technology out there, because there is. And some of it delivers beyond expectations. It's just that many people—both developers and users of technology—have become frustrated.

One of the problems is excessive complexity, according to Richard LeFrancois, President of Equipment Maintenance Innovators in Littleton, CO. His experience with fleet management began in 1972 as a driver. His construction equipment experience includes sales of field support vehicles, material-handling cranes, off-road tire-handling equipment, and other products. He has represented a number of firms, from Iowa Mold & Tool to Mani-towoc. He uses his membership in the Geospatial Information and Technology Association to further study successful field-force automation, project management, and field-data acquisition.

Engineers at Cummins, Inc. consider QuickCheck a good example of simple technology delivering dependable performance.LeFrancois believes engineers have developed, and marketers have sold, new products without understanding end users' needs. "The original task comes down to gathering data. It's a tool not available to us in years past. Fleet managers can now see an event with a date/time/frequency stamped record. That information allows managers of modern equipment to make decisions immediately, usually before catastrophic failure occurs. In the past, managers often had to modify maintenance procedures when problems were diagnosed in a post-failure tear down.

"I need maybe six reports to tell me the condition of my fleet, and that will allow me to have 75 or 80% uptime," said LeFrancois. "Those might include operating hours, machine location and movement, systems pressures, and shock (data).

"So the engineers should say, 'Okay, Mr. Manager, tell us which six reports we're going to focus on. We'll give you the technology to automate the data collection process. You'll get your reports in a timely and accurate fashion.

Instead, claims LeFrancois, engineers bundle every trick in their repertoire into the product, resulting in too much complexity and too little reliability. "I know of an operation where it took three trips (to the customer's site) just to get the hardware to work, to get it stabilized, always working, impervious to shock and vibration."

LeFrancois believes the solution is to scale back on product and make better use of existing maintenance procedures. From his perspective, "it's cost effective to hire people to collect data as opposed to putting a data collector on each piece of equipment."

Use people instead of technology? Heresy. Yet the industry is re-examining the role people play in the deployment of technology.

Edward Calvesbert is one adherent to this approach of leveraging people, handheld technology, and existing business processes to reduce the cost of a fleet-management system. He is the FleetRanger Product Manager at GlobeRanger Corp. FleetRanger is an online fleet-management application that uses wireless technology to track assets and monitor their performance. An engineer by training, Calvesbert has a graduate degree in construction technology from Stanford University and extensive experience developing software solutions for the construction industry.

"Devices that are specific to an asset, that you buy and leave in place, cost millions of dollars to deploy across a fleet," said Calvesbert. "Examples include those systems traditionally offered by Qualcomm, MicroLogic, and Deere Trax. But if you centralize some of that on a handheld device, it's cheaper to gather the data and reap the value of that data through whatever software you use."

One weakness of most handhelds is their lack of ruggedization. "If you drop one from five or six feet," said Calvesbert, "you'll probably break it. They're relatively cheap so you can say, 'Well I'll just buy another one,' but I don't think that's a long-term solution."

Despite their fragility, handhelds are becoming the tool of choice. Why? Low cost. Another advantage is portability, a benefit made even better with the recent emergence of Java 2 Micro Edition (J2ME). "With J2ME you can have something running on a Palm, on a Nextel, on a RIM, on a Windows CE device," said Calvesbert. "Everybody supports Java. That allows a company to develop a technology and then deploy it across a whole lot of different platforms."

Cummins' QuickCheck tool is designed for handhelds. The original (and the soon-to-be-introduced Version 2) relied on the Palm OS. "Palms are lightweight, portable, and the price point is where people want it," said Andy Pajakowski, Engineering Manager, Industrial Electronics at Cummins, Inc. Part of his role is to identify and implement current and upcoming off-the-shelf technologies for use by Cummins' industrial engine markets. He also helps in mapping an implementation strategy for future development of information-based products. With QuickCheck, Pajakowski was responsible for identifying customers' requirements for a data-link diagnostic tool based on the Palm OS.

Palms now come with 8 Mb of memory. With QuickCheck's tiny application file (only 300 kb), there's room for data from 300 trucks.

"With Palm OS," said Pajakowski, "you can write very small applications that do very focused things for the end user. We keep QuickCheck focused."

There are three basic elements in QuickCheck. One stores unit information, such as make, model, and serial number. Another stores fault codes with a description. A third stores trip information, such as hours or mileage and fuel. Version 2 will be a dual-mode unit, with both J1587 and J1939 connections. (Version 1 had only a J1587 interface.) That version will also be available to the consumer market for owners of Cummins-powered Dodge pickup trucks.

A single tool capable of tweaking ECMs would be appealing, but Pajakowski doesn't think it's likely to happen. "When you start to manipulate data, you get into proprietary protocols. Each manufacturer has different ways of changing data, and they don't tell anybody how to do that. It's a trade secret."

Lack of standardization may be the biggest anchor holding back the integration of technology. "End users, equipment designers, and service technicians want universal schemes," said CCG's Nelson, "but systems manufacturers don't want that. It's hard to get on anybody's horse when nobody wants to be universal."

There are glimmers of hope. "We have our own proprietary data link," said Mike Gatz, Manager of Information Products for Caterpillar, Inc.'s Electronic and Electrical Systems Division. "We began putting the Cat Datalink on machines in the mid-1980s because other standards that met our requirements did not exist. But the rapid adoption of electronics onto an ever-growing machine product line has also driven a significant growth in data link traffic. That growth in traffic can require faster data rates than our proprietary data link will comfortably accommodate."

doc2 4That growth and the push for universal standards have driven Cat to adopt other industry data-link protocols. "The European marketplace is driving the use of standards to simplify the integration of components from a variety of suppliers. In addition, the U.S. is requiring adoption of standard data links to support the monitoring of engine emissions through service tools and other test devices. These standards will help in diagnosing engines that may not be meeting emissions requirements."

Gatz says one of the impediments to global standards is the array of applications for equipment. Consider the needs of a motor grader, a fishing trawler, and a genset. The demands of each are quite different. The only need they have in common is basic diagnostics, and that's the only aspect seeing any real standardization.

Calvesbert uses a backhoe and a crane in his example. "You want to use the same system for the GPS devices on a backhoe and the telemetry devices on a crane. The software should be independent of the actual devices being used to capture the different types of data. Regardless of whether you're using an automated tracking device, a handheld device, or a combination of both, the software should be able to normalize the different types of information and generate the appropriate exception reports."

Dissimilarity isn't the only hurdle in the integration of technology across the off-highway industry. While pockets of technology exist, the construction equipment market is too small to justify the investment by product developers necessary to integrate everything from site plans to machine control to preventive maintenance.

The one exception is GPS. In that case, the construction industry is in the driver's seat.doc2 5

"You can get GPS at several levels of accuracy," said Chuck Schaidle, Manager of Product Management for Caterpillar's Electronic and Electrical Systems Division. At its broadest level, GPS offers general location—enough to know if the machine has been taken outside its geofence. At the range of several meters, a service truck can locate the equipment. At the meter and submeter range, agriculture and mining applications are available to minimize loss and waste. At the centimeter level, machine control is possible.

With real-time kinematic GPS, accuracy to 20 mm (0.8 in) in X and Y and 40 mm (1.6 in) in Z are consistently achievable. Grade stakes and chain crews are replaced by digital site plans, and changes are just a download away. Getting survey crews off the site improves both productivity and safety.

"The automotive industry doesn't need centimeter accuracy," said Schaidle. "Neither does the airline industry. But clearly we do. As our volumes start picking up, we'll become the driver of that technology."

The next role for GPS will be subterranean mapping of utilities, according to Gatz.


Necessary for the integration of technology is reliable, affordable wireless connectivity. The cell, satellite, and pager networks are, for the most part, built out. Advances now rely on more subtle developments.

Ki Chun is Chief Technical Officer of Luxul Corp., which markets proprietary antennas and builds metropolitan area networks (MAN), wireless local area networks (WLAN), and high-speed point-to-point (P2P) data-back haul networks. Chun explains that his firm is attacking the problem on three sides: antenna technology, radio modulation, and routing.

The typical antenna propagates sine waves, which, when reflected from flat surfaces (such as buildings), can cancel each other out because sine waves are used in linear polarization.

"The polarization we use is like a corkscrew," said Chun. "It's a circular polarization." He uses a corollary of the right-hand rule for determining the direction of a magnetic field based on current flow to explain how the technology works. He says both NASA and the U.S. military use circular polarization. "The advantage is that you minimize multipath effects, which are the greatest attenuators of RF signal."

What multipath signal remains can be cleared up inside the radio. "We look at orthogonal signals to see if they're the same signal," said Chun. "If they are, we combine them. You get one large, clear signal instead of a number of orthogonal, unclear signals. The key is to have a very, very high signal-to-noise ratio. In other words, we use a modulation scheme instead of an amplification scheme."

Chun was less open about Luxul's nascent plans for routing signals. The core concept is the use of clients to relay signals. "If you can build a ubiquitous, general autonomous network by linking clients, you don't have to put up this monolith of cell towers. All you have to do is get subscribers and density."

Such innovations, along with others (such as Chun's idea of an adaptive antenna with artificial intelligence), will be required to propel wireless connectivity from concept to reality.

Business competency

"We've got to understand what we're trying to do before we can go through all this technology," said LeFrancois. "I had an interview with a truck equipment manufacturer. I said, 'Okay, Mr. CEO, you're a $110-million a year company. You build this equipment. What six reports do you want to see?' He couldn't tell me. He didn't know. And I thought, 'We have a long way to go here.'"

Nelson agrees. "If it's not a good operation in the first place, our software is not going to do a thing. Not even the best software can correct for a lousy manager."

Because they are less familiar with emerging technology than they are with, say, tires and oil, managers often make unwise purchasing decisions regarding new technology. They can't always detect the marketing distortions. "People distort the capabilities of their technologies," said Nelson. "They hope their device will work, when they should be confirming that it does."

Another frequent oversight is the true cost of implementing new technology. "The more technology you throw at anything, the harder it becomes to see the direct correlation between technology expense and actual benefit," said Calvesbert. "It's not just the technology, but the installation, training, and monitoring. If you can't quantify how the technology is actually saving you time or money in the short term, then there's something wrong."

"Nobody understands that the cost (of the technology) is just half the true cost," said Nelson. "The other half is training your people, getting them to accept change, and that can be painful."

There are several reasons why it takes awhile for new technologies to roll out in a big way. "The challenge with service is for dealerships to alter their processes to keep up with technology," said Jim Wagner, Senior Project Engineer with Caterpillar's Service and Parts Operations Division. "We have this infrastructure built on traditional processes. As we develop new technologies, we have to institutionalize those changes into the service network."

Nelson believes safety and liability have inhibited adoption. "What is the reliability of the technology? What is the manufacturer's liability if something fails? Equipment is not one of those sexy things like HDTV you can try out and if something fails, no one gets hurt." The result of all this concern, she said, is that companies tend "to see much slower deployment than originally anticipated."

The future

No matter how advanced the processes become, they are useless unless they help shape a business owner's plans. "One challenge will be to turn data into knowledge and information that an operator or manager can actually use," said Calvesbert. "Right now the software is lagging. The information isn't always being represented visually or richly enough for users to make decisions from that information."

Cummins is also working on this angle, according to David Spear, Mining Marketing Director. The infrastructure is in place, including such sophisticated sensors as exhaust gas temperature probes. "If you see a cylinder start
to go cold, that's a good indication that something's wrong," he said. However, on an 18-cylinder engine, the operator won't feel one of those cylinders drop out, so the program is better at diagnostics. Maybe too good.

"With our Mining Gateway program, we're downloading some 35 variables every day," said Spear. Those variables include exhaust port temperature, intake manifold temperature and pressure, engine coolant temperature and pressure, engine speed, oil filter pressure differential, and fuel inlet restriction. The program also provides a duty cycle map.

"We talked to customers and found out that if things are going good, they don't want all that data," Spear continued. "So we're working on an intelligent system. It's software that will mimic a human's ability to look at a bucket of information and spot problems." Ideally the software will ferret out trouble, alert the dealer or service technician, and the whole diagnostic and repair sequence will be nearly invisible to the mine operator.

Improved service is also Wagner's goal. He sees three areas of improvement. First, better planning will put the technician on the site with the right tools and parts to do the repair correctly the first time. Second, the technician will be able to exchange service information from the site. Third, the machine would tell the operator whether a fault requires immediate shutdown and repair or if the machine can finish the shift at reduced power.

Prudence, not avoidance

So does all this mean end users should shy away from technological improvements in equipment? They probably couldn't, even if they wanted to. Technology is everywhere. Besides, successful companies know how to use technology to their advantage.

"It's not going to go away," said Nelson. "The trick is in discovering how it can make your life better."


Reprint from "SAE Off-Highway Engineering - September 2001"