Stopping the Stops: How an Automotive Parts Manufacturer Used Consulting and Analytics to Identify the Real Causes of Unplanned Downtime Across Its Horizontal Machining Centre Fleet

Reactive maintenance in a high-volume automotive parts plant is a compounding problem. Each unplanned stop generates a direct cost in lost production, a labour cost in emergency response, and a relationship cost with the customer whose just-in-time delivery schedule has just been disrupted.

The client’s maintenance team was experienced and had strong intuitions about which machines were most troublesome and which failure modes were most common. Those intuitions had been built from years of hands-on experience but had never been tested against the full maintenance log data.

The plant manager was also facing a capacity problem. The maintenance budget was not sufficient to implement comprehensive preventive maintenance across all sixteen machines simultaneously, which meant any improvement programme needed to be targeted at the machines and failure modes that would produce the largest downtime reduction for the resources available.

Key Challenges

• No formal analysis of the plant’s two-year maintenance log to identify the actual distribution of downtime causes, machines, and timing patterns across the fleet.
• A maintenance team with strong intuitions about failure modes built from experience but never validated against aggregated data.
• Coolant system incidents logged individually as minor events that had never been aggregated into a pattern visible in the maintenance team’s experience-based view.
• No leading indicator analysis examining whether machine parameter data contained early warning signals preceding the most costly failure modes.
• A maintenance budget insufficient for comprehensive fleet-wide preventive maintenance, requiring targeted allocation to the highest-impact machines and failure modes.
• Customer delivery penalties accumulating from unplanned downtime running above target for two consecutive years.

Reactive maintenance feels like it is responding to the problem, but it is almost always responding to the symptom. The actual problem, the failure mode that keeps recurring, is buried in the maintenance log pattern that nobody has yet aggregated into a diagnostic picture. In a high-volume automotive plant, that picture is usually more concentrated than the maintenance team expects, and the right response is almost always more targeted than a general preventive maintenance programme.