Switching Suppliers Does Not Fix a Procurement Process Problem
Quote from chief_editor on June 27, 2026, 5:30 pmWhen equipment quality problems recur across multiple suppliers, the problem is often in the procurement process—specification, evaluation, or monitoring—not in the specific suppliers being used.
The bearing housing castings had been producing excess porosity in the pressure-boundary walls. The buyer—a pump manufacturer sourcing castings from China—had used the same Hebei foundry for three years. After the porosity issues emerged in two consecutive batches, the decision was made to switch to a different foundry in Shandong. The Hebei foundry was removed from the approved vendor list.
The Shandong foundry produced the first batch. Porosity was present in the same locations as in the Hebei batches.
A metallurgical review commissioned after the second foundry's failure identified the root cause: the casting design drawings specified wall thicknesses in the pressure-boundary sections that created a thermal gradient during solidification that was known in foundry engineering practice to produce shrinkage porosity in iron alloys. The design drawings—which both foundries had received as the basis for production—specified a geometry that was inherently prone to the defect being attributed to foundry quality.
Neither foundry had flagged this issue. Neither had been asked to review the design for castability during the qualification process. Both had produced the geometry as specified. Both had produced porosity in the locations where the geometry created the conditions for it.
The problem was not with the Hebei foundry. It was not with the Shandong foundry. It was in the casting design.
When Supplier Changes Replicate the Problem
A quality failure that recurs across multiple suppliers is a reliable indicator that the source of the problem is not supplier-specific. Supplier-specific problems—a foundry with inadequate process control, a machining operation with worn tooling, a fabricator with undertrained welders—produce failures that are resolved when the supplier is changed for one with adequate capability. The pattern is: one supplier with problems, resolution through supplier change, new supplier without the problem.
When the pattern is: one supplier with problems, supplier change, new supplier with the same problem—the source of the problem is almost certainly in the procurement input rather than in the specific supplier. The procurement input includes: the technical specification, the design drawings, the material specifications, and the test and inspection requirements. If all of these inputs are consistent across suppliers, and the failure is consistent across suppliers, the failure's cause is embedded in the inputs.
This causal pattern is common in industrial equipment component procurement and is consistently underdiagnosed. The response to a quality failure is almost always first-order: identify which supplier produced the defective parts and replace them with a supplier who is expected to produce better parts. The second-order question—why did the defect occur, and is the procurement specification contributing to its occurrence?—is less frequently asked, partly because it is more difficult to diagnose and partly because it implies that the problem originates within the buyer's own engineering function rather than with the external supplier.
For the bearing housing castings, the second-order question was not asked after the Hebei foundry failure. It was asked only after the Shandong foundry produced the same pattern—at which point the evidence was strong enough to warrant a metallurgical review.
Identifying Specification-Originated Quality Problems
A quality failure that has a specification origin is distinguished from a supplier-capability failure by several characteristics.
The failure occurs in the same location across multiple parts and across multiple suppliers. Supplier-capability failures are distributed across locations that vary with the specific production variation. Specification-originated failures concentrate where the specification creates the problematic condition—in this case, the geometric features that caused thermal gradient porosity were consistent across both foundries because the geometry was consistent.
The failure occurs on the specific features that are most demanding in the specification. If the failure concentrates on the tightest tolerance, the most challenging geometry, the most demanding surface finish, or the highest performance requirement, the specification may be requiring more than the available manufacturing capability can reliably deliver—and changing suppliers selects among capable suppliers without changing what is being demanded.
The failure does not respond to supplier quality improvement actions. When a supplier implements corrective actions—adjusting process parameters, improving incoming material inspection, increasing sampling rates—and the failure rate does not change, the corrective actions are not addressing the root cause. This is a signal that the root cause is outside the supplier's control.
For the casting problem, the design change was straightforward once the root cause was identified: adding a boss at the thin-section location to increase the thermal mass and eliminate the shrinkage gradient. The modification added $12 per casting in machining cost. It resolved the porosity issue across both foundries with the existing qualification without any foundry process changes required.
The three-year problem had been solved by a twelve-dollar design modification that no one had looked for because the investigation had consistently started and ended with the supplier rather than with the specification.
Quality problems that recur across supplier changes are telling the buyer something about their procurement process that is more valuable than any single supplier evaluation. Listening to the pattern is more useful than looking for a better supplier.
When equipment quality problems recur across multiple suppliers, the problem is often in the procurement process—specification, evaluation, or monitoring—not in the specific suppliers being used.
The bearing housing castings had been producing excess porosity in the pressure-boundary walls. The buyer—a pump manufacturer sourcing castings from China—had used the same Hebei foundry for three years. After the porosity issues emerged in two consecutive batches, the decision was made to switch to a different foundry in Shandong. The Hebei foundry was removed from the approved vendor list.
The Shandong foundry produced the first batch. Porosity was present in the same locations as in the Hebei batches.
A metallurgical review commissioned after the second foundry's failure identified the root cause: the casting design drawings specified wall thicknesses in the pressure-boundary sections that created a thermal gradient during solidification that was known in foundry engineering practice to produce shrinkage porosity in iron alloys. The design drawings—which both foundries had received as the basis for production—specified a geometry that was inherently prone to the defect being attributed to foundry quality.
Neither foundry had flagged this issue. Neither had been asked to review the design for castability during the qualification process. Both had produced the geometry as specified. Both had produced porosity in the locations where the geometry created the conditions for it.
The problem was not with the Hebei foundry. It was not with the Shandong foundry. It was in the casting design.
When Supplier Changes Replicate the Problem
A quality failure that recurs across multiple suppliers is a reliable indicator that the source of the problem is not supplier-specific. Supplier-specific problems—a foundry with inadequate process control, a machining operation with worn tooling, a fabricator with undertrained welders—produce failures that are resolved when the supplier is changed for one with adequate capability. The pattern is: one supplier with problems, resolution through supplier change, new supplier without the problem.
When the pattern is: one supplier with problems, supplier change, new supplier with the same problem—the source of the problem is almost certainly in the procurement input rather than in the specific supplier. The procurement input includes: the technical specification, the design drawings, the material specifications, and the test and inspection requirements. If all of these inputs are consistent across suppliers, and the failure is consistent across suppliers, the failure's cause is embedded in the inputs.
This causal pattern is common in industrial equipment component procurement and is consistently underdiagnosed. The response to a quality failure is almost always first-order: identify which supplier produced the defective parts and replace them with a supplier who is expected to produce better parts. The second-order question—why did the defect occur, and is the procurement specification contributing to its occurrence?—is less frequently asked, partly because it is more difficult to diagnose and partly because it implies that the problem originates within the buyer's own engineering function rather than with the external supplier.
For the bearing housing castings, the second-order question was not asked after the Hebei foundry failure. It was asked only after the Shandong foundry produced the same pattern—at which point the evidence was strong enough to warrant a metallurgical review.
Identifying Specification-Originated Quality Problems
A quality failure that has a specification origin is distinguished from a supplier-capability failure by several characteristics.
The failure occurs in the same location across multiple parts and across multiple suppliers. Supplier-capability failures are distributed across locations that vary with the specific production variation. Specification-originated failures concentrate where the specification creates the problematic condition—in this case, the geometric features that caused thermal gradient porosity were consistent across both foundries because the geometry was consistent.
The failure occurs on the specific features that are most demanding in the specification. If the failure concentrates on the tightest tolerance, the most challenging geometry, the most demanding surface finish, or the highest performance requirement, the specification may be requiring more than the available manufacturing capability can reliably deliver—and changing suppliers selects among capable suppliers without changing what is being demanded.
The failure does not respond to supplier quality improvement actions. When a supplier implements corrective actions—adjusting process parameters, improving incoming material inspection, increasing sampling rates—and the failure rate does not change, the corrective actions are not addressing the root cause. This is a signal that the root cause is outside the supplier's control.
For the casting problem, the design change was straightforward once the root cause was identified: adding a boss at the thin-section location to increase the thermal mass and eliminate the shrinkage gradient. The modification added $12 per casting in machining cost. It resolved the porosity issue across both foundries with the existing qualification without any foundry process changes required.
The three-year problem had been solved by a twelve-dollar design modification that no one had looked for because the investigation had consistently started and ended with the supplier rather than with the specification.
Quality problems that recur across supplier changes are telling the buyer something about their procurement process that is more valuable than any single supplier evaluation. Listening to the pattern is more useful than looking for a better supplier.