Case 20: Rondell Data Corporation John A. Seeger Bentley College "Damn it, he's done it again!" Frank Forbus threw the stack of prints and specifications down on his desk in disgust. The Model 802 wide-band modulator, released for production the previous Thursday, had just come back to Frank's Engineering Services Department with a caustic note that began, "This one can't be produced, either. . . ." It was the fourth time Production had kicked the design back. Frank Forbus, director of engineering for Rondell Data Corp., was normally a quiet man. But the Model 802 was stretching his patience; it was beginning to look just like other new products that had hit delays and problems in the
That would give us a real jump on the competition."

Four days after that meeting, Frank found that two of his key people on the 802 design had been called to Production for emergency consultation on a bug found in final assembly: two halves of a new data transmission interface wouldn't fit together because recent changes in the front end required a different chassis design for the back end. Another week later, Doc Reeves walked into Frank's office, proud as a new parent, with the new filter design. "This won't affect the other modules of the 802 much," Doc had said, "Look, it takes three new cards, a few connectors, some changes in the wiring harness, and some new shielding, and that's all." Frank had tried to resist the last-minute design changes, but Bill Hunt had stood firm. With a lot of overtime by the engineers and draftsmen, Engineering Services should still be able to finish the prints in time. Two engineers and three draftsmen went onto 12-hour days to get the 802 ready, but the prints were still five days late reaching Dave Schwab. Two days later, the prints came back to Frank, heavily annotated in red. Schwab had worked all day Saturday to review the job, and had found more than a dozen discrepancies in the printsmost of them caused by the new filter design and insufficient checking time before release.