Throughput recovery in a two-plant industrial group
Audited under the Markham Verification Standard at month 9; run-rate re-tested at month 12. Full audit trail available under NDA.
The situation
A two-plant industrial group had order books it could not ship. Lead times had stretched from three weeks to seven, overtime had become structural rather than occasional, and the board was reviewing a $14M capacity expansion as the obvious answer to all of it. The capex case was internally persuasive and externally unexamined.
The group asked for a second opinion on the expansion. We thought the more useful question was where the existing capacity was already going, because at nameplate the two plants between them held 26% more capacity than the order book required.
What the diagnostic found
Four weeks of measurement at the line located the loss with some precision. Two bottleneck work centres were being starved by upstream sequencing decisions that were made daily, informally, and differently on every shift. Rework at those centres consumed 19% of their hours. Priority orders reset the schedule a median of six times a week, and the order that got expedited was usually the one whose customer had escalated loudest rather than the one whose delay cost the group the most.
None of this was a capacity problem. It was a decision problem wearing a capacity costume. There was no agreed sequencing rule, no fixed forum with the standing to hold one, and no baseline anyone trusted.
How Markham helped
The work ran in three phases over nine months and asked for no capital. Phase one was the constraint map. We located and priced the losses line by line and froze a month-0 baseline that both plant directors signed. Phase two redesigned the cadence, which meant one sequencing rule for the bottleneck centres, a daily fifteen-minute schedule forum with the authority to hold that rule, and a weekly operating review of run-rate against the baseline.
Phase three was the hold. Three months of the new rhythm under oversight, with every expedite request routed through the forum and priced openly. The loudest-voice escalation channel closed because it stopped working. What the plants had to supply was attendance, because a daily forum that the people who can actually commit to a sequence do not attend is only another meeting with a shorter agenda. The hold is the phase clients ask us to shorten, and it is the one we will not.
Impact in detail
Baselines frozen and audited before phase one; month-9 readings verified under MKM-F-003 and re-tested at month 12.
What we took from it
The capex case was real arithmetic on top of a false premise. Mapping the constraint before pricing the expansion saved $14M. We would not generalise from a single group to a rule about capital projects, but we now ask where the existing capacity is going before we read anyone else’s expansion business case.
The binding constraint was a daily decision, not a machine. Fixing who sequences the bottleneck outperformed adding capacity to it.
A run-rate that is not reviewed weekly decays. The hold phase is the least glamorous third of the work and the only reason the result outlasted our involvement.