The safety engineer, a woman named Carla Benton, began with Press Seven. She studied the machine's design, interviewed technicians, and examined archived diagnostic data. The defect proved more complicated than the broken relay Pete had suspected. A contact in the light-curtain circuit was failing intermittently, and the diagnostic jumper had bypassed a signal that should have prevented normal operation after testing. In certain conditions the press could resume a cycle without the full confirmation required by its safeguarding system. Carla immediately ordered the machine to remain isolated until the circuit was rebuilt and validated. She also checked the other presses using the same controller, finding two that needed corrective maintenance before they could be returned to routine service.
Pete was interviewed for nearly four hours. He produced work orders showing that he had flagged the jumper as temporary and recommended against restarting the press until it was removed. The maintenance computer showed that his warning had been attached to the work order, then obscured when a manager closed the job using a different completion code. The closing entry was not in Pete's account. It belonged to a production manager named Gregory Vale, who had been under direct pressure to recover lost capacity. Gregory initially said the closure had been an administrative mistake. Under questioning, he admitted he had authorized a restart after a brief functional test because he believed the remaining fault was unlikely to cause injury.
Carla did not need to speculate about his motives. She documented the gap between the approved maintenance procedure and the actual restart decision. Gregory had no authority to override the safety hold, and the process should have required a second qualified signoff. The company's system allowed one manager to close a maintenance ticket and release equipment without independent confirmation. That weakness did not excuse Gregory's decision, but it made the risk broader than one employee. A different manager with the same access could have made the same mistake the following week.
The review team interviewed Marla next. She described the near miss when she had pulled a young operator's arm away from the die, the complaint she filed afterward, and the embarrassment of being told she was resistant to operational needs. She did not describe herself as a perfect worker. She admitted she had missed two shifts when her mother was in the hospital and had sometimes argued sharply with supervisors. But she challenged the idea that those matters justified a sudden finding of chronic poor performance after years of strong evaluations. She asked the investigators to speak with the people she trained, including those who disagreed with her about everything except whether she knew the job.
One of those people was Rachel Knox, a skilled assembler who had once competed with Marla for a lead position. Rachel Knox had been promoted while Marla remained an operator, and their relationship had not always been easy. Yet when the investigator asked about Marla's work, Rachel said she was one of the strongest troubleshooters in the department. "She can be stubborn," Rachel Knox admitted. "Usually because she's right and somebody wants her to stop saying it. I've been annoyed with her plenty. That doesn't mean she deserves to be called incompetent."