DENVER - After observing troubled tests on a new baggage system at Denver International Airport yesterday, a United Airlines executive said the city should delay the airport's opening a fifth time.

"Testing hasn't gone well at all. Nothing we've seen today suggests it's going to be ready to go on [May] 15th," said United customer service manager Stephen Speers.

The $4 billion-plus airport is scheduled to open in two weeks.

"Without the bag system, a hub airport won't run," Speers said. "Our view is when it's finished, it's going to be the best airport in the country, and the bag system will be the best in the world, but we shouldn't open before we're ready," the United official said in an interview with The Bond Buyer.

A city official said, however, that airline officials and other should not jump to conclusions about the troubled baggage system.

"I need to emphasize that we have three days of testing," said Denver public works director Mike Musgrave. "We hope that every day we build on the previous day. We're still looking forward to improved performance."

Separately, United this week wrote a formal letter to the city requesting another delay. But it is unclear who would pay for the approximately $15 million a month cost for another delay. United and other airlines have picked up the tab in the past.

The city is drafting a response to United's letter.

Musgrave and others, including BAE Automated Systems Inc., the designer and builder of the airport's baggage system, have pleaded with the airlines to defer passing judgment until this weekend's tests are complete.

Tests yesterday on the $193 million automated baggage system at Denver International continue to show jammed cars, software bugs, and other problems.

A test scheduled for yesterday morning on the baggage circuit that serves United Airlines' started late and was aborted after only about 100 of 900 bags reached their destinations on United's Concourse B, according to city officials and airline sources at the test site. I was the first of several days of testing.

A separate test on the circuit system that serves Continental Airlines' was more successful, but complete information on the number of bags successfully routed to the Concourse from the terminal was still being compiled at deadline.

Musgrave said the Continental test was more successful than United's. Ten minutes after Continental's test began, however, jams caused the system to be shut down for a half hour, before being restarted.

"We did experience some jams, but they seemed to cleared quickly," Musgrave said.

United's test did not fare as well. Five minutes after the test began simultaneously with Continental's, a jammed cart on Concourse B closed down the system. It was cleared, but other jams followed that one.

When the system kicked on again 45 minutes later, the baggage carts that were supposed to pick up luggage in the terminal building circled around the building without stopping. In other cases, the proper number of empty carts did not go where to they were supposed to, a continuation of the chronic "empty-car management' problem that is caused by programming errors.

Regarding the United test, Musgrave said, "We experienced more jams than we'd have liked to seen. We're going to re-group and start again this afternoon." At press time, the circuit had not yet been started up.

Another problem is allowing the airlines enough time to train their employees on the system. Contractually, the city is must give two months to train on the system.

Speers said United could train its people in two weeks.

"A month would be ideal, but we're willing to go the extra mile," he said.

The testing schedule for late Thursday and Friday called for 7,000 bags to be shuttled from the A and B concourses, which handle 80% of the airport's traffic. But the airlines produced only 4,000 bags for testing by Thursday.

Subscribe Now

Access to authoritative analysis and perspective and our data-driven report series.

14-Day Free Trial

No credit card required. Complete access to articles, breaking news and industry data.