In the annals of project management literature, there are few stories as compelling as that of the Denver International Airport (DIA). DIA – which was to replace Stapleton International Airport – was scheduled to open in October of 1994 with the construction budget being set at $2B. It eventually opened 16 months late, ultimately costing $4.8B, an almost 250% increase. Maintaining the empty airport and interest charges on construction loans cost the city of Denver $1.1M per day throughout the delay. The main reason pointed to for the delay was scope changes requested by United Airlines. (Italics and boldface are mine.)
If you want to read the full story, there are many case studies on the Internet. I’ve quoted from a few below. (Tellingly, one of them is subtitled “An illustration of ineffectual decision making.”1) If you’re pressed for time, consider this an executive summary. For the record, DIA is, of course, open. I’ve flown into it a number of times in the past few years and it’s quite nice. So this is not the story of a project that was abandoned. It is the story of a technology project run amok, which became much harder and much more expensive than it ever had to. There are far too many reasons underlying this fiasco to fit in this post. I will focus here on three, summarizing some others below:
Design complexity/Scope changes
The baggage handling capabilities of the system – or lack thereof – are fundamental to understanding the initial massive failure. They represent the vast majority of the scope changes requested by the airlines. In short, everyone involved underestimated the complexity of the system. But even when it was pointed out that A) the system was too complex and B) no bidder could build it on time, the decision was made to go ahead anyway. (See below, Project Management Expertise.) And even if the complexity issue had been factored in – and solved- it never occurred to anyone to update the original schedule, much less budget. (Scope Creep).
I will now here quote project management guru Harold Kerzner on the complexity of this proposed “never-been-done-before” baggage handling system: “The system would contain 100 computers, 56 laser scanners, conveyor belts, and thousands of motors. … It would contain 400 fiberglass carts, each carrying a single suitcase through 22 miles (!) of steel tracks. Operating at 20 MPH, it could deliver 60,000 bags per hour from dozens of gates. ..To illustrate the complexity of the situation, consider 4000 taxicabs in a major city, all without drivers, being controlled by a computer through the streets of a city.”2 (Google working on something like this?) For the record, there were some 2100 design changes.
In April 1994, the airport arranged for a demonstration of what is likely the most complex – luggage handling system ever built. (Alas, without informing BAE Systems, who designed it). Result? In a nutshell, bags were crushed, clothes disgorged, and public humiliation was had. Jay Leno, then of the Tonight Show, had a field day with it. (To be continued).