Strategic CIO // Executive Insights & Innovation
News
8/3/2012
11:05 AM
Connect Directly
LinkedIn
Twitter
RSS
E-Mail
50%
50%

Why Union Pacific Builds Its Own Tech

The railroad often finds off-the-shelf software and hardware is too expensive and isn't tuned to its needs.

In an era when most IT organizations look to buy off-the-shelf software, avoid doing custom coding, and allow the use of more mass-market consumer technology than they ever thought possible, Union Pacific CIO Lynden Tennison is an unapologetic believer in building rather than always buying the hardware, software, and systems that underpin its $20 billion railroad shipping business. And while it's at it, UP generates $35 million to $40 million in revenue a year selling or licensing some of those technology innovations to other companies, including rivals, creating a small profit center within its $300 million-a-year IT cost center.

UP, the nation's largest railroad company, operating west of Chicago and New Orleans, develops and builds much of its own IT partly out of necessity. The railroad shipping industry's specialized and often complex technology needs don't add up to a lucrative enough market to attract the major tech vendors. And the boutique vendors that try to fill those specialized needs are often one bad quarter away from bankruptcy, Tennison says, or they charge exorbitant prices because they're not producing at large scale and lack adequate competition. So UP is always looking for in-house IT alternatives.

While Tennison jokes that UP isn't likely to get into the general ledger software business anytime soon, it did develop its own supply chain applications, as the likes of SAP and Oracle couldn't meet its stringent performance requirements (it will tolerate only 1.5 hours of downtime a month for system upgrades and changes). UP even develops its own firewalls; develops its own service-oriented architecture; builds its own locomotive radios, networking consoles, and antenna farms; owns its own radio spectrum and trackside optical fiber (34,000 miles of it); and builds its own microwave towers (700 of them).

More ambitiously, UP acquired a small Michigan-based gaming outfit several years ago to develop virtual reality programs--complete with avatars and precise 3-D renderings of actual rail yards and trains--for internal training of RCOs (remote control operators), conductors, car inspectors, and other personnel. The company is now looking to sell that software to other railroad companies, as well as to construction, mining, and energy companies, under its PS Technology unit.

The mother of all UP IT projects is NetControl, a Linux-based transportation management system to replace the mainframe-based system the company rolled out in the late 1960s. About 270 people worldwide are working on NetControl, which Tennison calls "our family jewels." The $200 million system, which is rolled out in increments, is about halfway done and due to be fully operational in 2017.

UP's current transportation management system is used mostly for taking customer orders, scheduling train operations, tracking what's going on in train cars, and monitoring railroad traffic. It's being updated to a Web platform, letting customers access it from mobile devices, place more accurate orders, and see the planned travel itinerary for their shipments. With NetControl, UC will be able to monitor terminal and network use over time and forecast imbalances in the supply and demand of crews or locomotive power. Automated action- or alert-driven workflows can recommend optimal solutions.

One internal benefit of moving to NetControl is that more technical expertise is available to support a Web versus a mainframe system, Tennison says. UP is also looking to sell NetControl to other railroad companies.

Tennison, the company's CIO since 2005, made his vendor bones running a for-profit UP service management application subsidiary called Nexterna from 1998 through 2001. Still, he says he sometimes wonders "whether we're dinosaurs" doing all of this custom work. "I second-guess myself a lot," he concedes, "but it doesn't feel like we're making bad decisions."


Comment  | 
Print  | 
More Insights
Comments
Newest First  |  Oldest First  |  Threaded View
Ohio Investigator
50%
50%
Ohio Investigator,
User Rank: Apprentice
8/9/2012 | 6:16:02 AM
re: Why Union Pacific Builds Its Own Tech
Kaiser---you have a very strong insight into an industry that constantly acts like it's still in either the 1800's or 1900's. They only do something having to do with safety when the FRA forces them to do it. Countries in Europe have had PTC for years. I read recently that even Russia has it on their trains. I always like to talk to people interested in railroads about why the RR's got rid of the caboose. It was a huge safety factor since the two crew members could see the entire train ahead of them. If they saw smoke or fire from a "hot box", they could dump the air and stop the train. If they saw a car vibrating from a broken wheel / derailment, they could stop the train. But, as usual, the RR's were totally focused on the Bottom Line and so, the train crew got sliced to 2 from 4 by eliminating the caboose. FRED was supposed to replace the caboose. Uh HUH!

We had a CSX incident some years ago here in Ohio where half the train was lost due to a decoupling and NOTHING detected it! The air brake system should have stopped the train, but there was a "kink" in the air hose and the system failed. So, the Engineer went his merry way, not knowing that about half of his train had separated and was derailed all over the track behind him. Some people living along the track saw the derailed cars and called 911. Isn't that just great? Of course, we had that "run-away" CSX train go 66 miles here in Ohio with no one in the locomotive! The Engineer didn't get fired (!) even though he rigged the throttle and jumped off to move a track switch (where was the Conductor, huh?) When he tried to get back on the loco, he slipped on the wet steps and had to let go as he was being dragged over the ties / ballast. That train got up to speeds of at least 60 mph. There has been a movie made about it: UNSTOPPABLE, 2010, starring Denzel Washington. It's an absolute miracle that no one was killed or injured by that real CSX "run-away" train and that it didn't derail. Where is Casey Jones when you need him?

SEE: http://csx-sucks.com/news/

Want to see something interesting about UP? Do a Google on the Alcorn case in Missouri in 1999. A very angry jury hit UP with something like $120 million in punitive damages! Yes, the judge knocked the amount down, but still..........
jclabby040
50%
50%
jclabby040,
User Rank: Apprentice
8/8/2012 | 6:42:35 PM
re: Why Union Pacific Builds Its Own Tech
Isn't Union Pacific the company that made the big splash in the tech mags a few years ago by telling the world how it was dumping its mainframe and moving exclusively to x86 architecture? And didn't they say they weren't going to worry much about virtualizing those servers? And didn't they say the migration and the complete rewrite of their code environment would cost between $150 million to $200 million?

Further, how much did it cost to develop all of the new code? How much was the re-tooling and how much was the cost to develop new skill sets? How much, over time, is it going to take to maintain that code? As for the 1.5 hours of downtime that they now allow in their service level requirements, how much downtime did they see when they were running their highly-available mainframe environment? How much money is being saved by having to build out, rewire, configure, and support a complete distributed network environment ? How much energy is being saved (I bet energy costs increased significantly...)?

Seems that I have to agree with the previous commenter: rail likes the position that they're in -- dominating small companies (whom I guess are the ones buying their software). But as for being risk adverse, I think Union Pacific took a huge risk dumping the mainframe for x86 architecture. Think of the security risks; the huge cost if the project failed; and the public embarrassment if we all find out that the mainframe is still there. To me, this shows Union Pacific is willing to take risks...

By the way, has Union Pacific finally moved off its mainframe or is it still there?
PJS880
50%
50%
PJS880,
User Rank: Apprentice
8/8/2012 | 1:36:37 PM
re: Why Union Pacific Builds Its Own Tech
It certainly appears that Union Pacific is the hot topic for the day! I am surprised to learn that UP is going with a web based system versus and enterprise system to manage their NetControl. I would think with all the in house IT that it already going on, I would think UP would want total control over managing NetControl. I also enjoyed reading that UP us earning millions based on some of the innovative technology they have already come up with. I would not worry if I were Tennyson it appears that his decisions are very good for business and that he is not making bad decisions at all.

Paul Sprague
InformationWeek Contributor
Kaiser Sose
50%
50%
Kaiser Sose,
User Rank: Apprentice
8/8/2012 | 1:25:17 PM
re: Why Union Pacific Builds Its Own Tech
Off the shelf software doesn't fit rail because the gang of 4 US-based railroads (UP, CSX, BNSF & NS) close themselves off to the world. Look at PTC. They had an opportunity to work with the industry and create something truly advanced. Instead, they teamed up, selected a handful of vendors, dictated the terms to the vendors and only pushed the technology as far as they were willing to invest. So, something that could have taken railroads into the digital age turned out to be a modern layer over 19th century technology.

I used to work for a tier 1 integrator that had ex railroad executives and was very interested in working with the rails. Yet, they pushed each and every offer aside in favor of going it alone out of fear that we'd also work with their competitors. For all of the talk that rail is a small market and it doesn't attract the large players is nonsense. The truth is the rails like the position they are in - dominating small companies and making small, risk averse advances in technology.

This market is ripe for new thinking.
The Business of Going Digital
The Business of Going Digital
Digital business isn't about changing code; it's about changing what legacy sales, distribution, customer service, and product groups do in the new digital age. It's about bringing big data analytics, mobile, social, marketing automation, cloud computing, and the app economy together to launch new products and services. We're seeing new titles in this digital revolution, new responsibilities, new business models, and major shifts in technology spending.
Register for InformationWeek Newsletters
White Papers
Current Issue
InformationWeek Tech Digest - July10, 2014
When selecting servers to support analytics, consider data center capacity, storage, and computational intensity.
Flash Poll
Video
Slideshows
Twitter Feed
InformationWeek Radio
Archived InformationWeek Radio
Join InformationWeek’s Lorna Garey and Mike Healey, president of Yeoman Technology Group, an engineering and research firm focused on maximizing technology investments, to discuss the right way to go digital.
Live Streaming Video
Everything You've Been Told About Mobility Is Wrong
Attend this video symposium with Sean Wisdom, Global Director of Mobility Solutions, and learn about how you can harness powerful new products to mobilize your business potential.