Focus on a small, stable set of waste KPIs
Material waste reduction is best measured with a small set of consistently defined KPIs rather than a long list of metrics. In regulated and mixed-vendor environments, the main challenge is stable definitions and reliable data capture across ERP, MES, and QMS, not inventing new indicators. Most plants benefit from tracking a core group: material yield, scrap rate, rework rate, and material cost of non-quality. These should be defined at the product, line, and plant levels to support both local problem solving and management reporting. Frequent redefinition of KPIs or ad hoc spreadsheets usually leads to confusion and weak trend analysis, which undermines improvement efforts.
Core rate KPIs: scrap, rework, and material yield
Scrap rate is typically measured as scrapped quantity divided by total input quantity for a given operation, line, or product within a defined time window. Rework rate is usually reworked quantity divided by total output quantity or total input quantity, depending on how your routing and MES handle rework loops. Material yield is often measured as good output quantity divided by total material input, sometimes normalized by standard bill of material quantities. In complex routings, you may need yield at key constraint operations rather than only at the final output. Whatever definitions you choose must be documented, controlled under change management, and applied consistently if you want to see real trends.
Cost-oriented KPIs: material cost of non-quality
To link waste reduction to business impact, you need at least one cost-based KPI such as material cost of non-quality. A common approach is to track the total material cost of scrap plus the incremental material consumed by rework, divided by total shipped product value or total material input. This requires accurate standard costs or, in some cases, actual costs from ERP, and traceable mapping from scrap and rework events to cost elements. In regulated environments, you must be explicit about whether you include only nonconforming material dispositioned as scrap, or also controlled destruction, expiry, and obsolescence. If your cost data are weak or delayed, start by measuring waste in physical units and gradually layer in cost once you can trust the integrations.
Quality and compliance-related waste KPIs
Some of the worst material waste comes from quality escapes, late-stage rejections, and controlled destruction, which may not appear in simple scrap summaries. You can track late scrap rate as the proportion of scrap generated after a defined process milestone, such as final inspection or test. Another useful KPI is batch or lot rejection rate, tied to material value, to show when entire units of work are lost due to systemic issues. In highly regulated plants, destruction due to expiry, storage conditions, or documentation failures can be measured as a separate KPI to highlight administrative and logistics-driven waste. These KPIs depend heavily on good lot traceability and alignment between MES, QMS, and warehouse systems; if that integration is fragile, trends may be more informative than absolute values.
Process and line-level KPIs: where the waste actually occurs
Plant-level waste KPIs must be decomposable to the operation and line level, or you will not be able to act on them. Operation-specific scrap and rework rates help identify which processes are generating the most loss, but they must be normalized correctly (e.g., per thousand units processed, not per shift, to avoid staffing bias). First-pass yield at critical operations is another useful KPI, defined as the percentage of units passing without rework or repair. Be cautious when aggregating across different part families or batch sizes, since this can mask local problems and confuse operators. In brownfield environments with limited MES coverage, you may need a hybrid approach where some operations are tracked in detail and others with periodic sampling or manual logs.
Inventory-related KPIs: expiry, obsolescence, and overconsumption
Material waste is not only about scrap on the line; expiry, obsolescence, and overconsumption in inventory can be equally significant. Expiry-related waste can be tracked as the percentage of inventory value written off due to shelf life or storage nonconformance during a period. Obsolescence can be measured similarly, tied to engineering changes or program terminations that leave materials unusable. Overconsumption can be defined as actual material usage versus standard bill of material across a period, with differences investigated to distinguish true process loss from data or configuration issues. These KPIs rely on accurate lot dating, controlled engineering changes, and disciplined inventory transactions; where those are weak, expect substantial reconciliation effort and uncertainty in the numbers.
Data and system constraints when defining waste KPIs
In mixed MES/ERP/QMS landscapes, you may not be able to implement all of these KPIs reliably at once. Some plants can capture scrap at the operation and lot level, but not reliably associate cost without manual mapping in ERP. Others have good cost visibility but poor routing-level yield data, making it hard to localize problems. In validation-heavy environments, any new KPI that depends on system changes or new integrations will require formal change control and potentially revalidation. It is often more realistic to start with a minimal, clearly defined KPI set that current systems can support, then incrementally refine definitions as integrations and data quality improve, rather than attempting a comprehensive redesign.
Choosing and governing your KPI set
When deciding which waste KPIs to adopt, select a small number that you can calculate consistently today, and that you can trace back to actionable process levers. Document each KPI’s exact definition, data sources, exclusions, and owner, and manage changes to those definitions under formal change control so trends remain meaningful. Align KPIs with existing continuous improvement practices so that value stream mapping, root cause analysis, and corrective actions naturally use the same numbers. Avoid designing KPIs that imply full system replacement just to calculate them, especially in aerospace-grade or similar contexts where qualification and validation costs are high. Over time, you can extend the KPI set as you gain better integration, but stable, trusted metrics are far more useful than a large, shifting dashboard of approximate figures.