Knowledge Centre​

Condition-based maintenance is a decision rule before it is a technology. This article sets out six rules that decide whether the programme pays for itself: filter assets by consequence rather than by how easy they are to instrument; name the failure mode behind every alert; set alert and alarm thresholds from a healthy baseline; keep the monitoring interval inside the P-F interval; reserve the part when the diagnosis is confirmed; and verify that each avoided failure was actually avoided. Drawing on the DOE O&M Best Practices Guide, ISO 17359 and a documented chemicals-plant condition monitoring case, it shows why faster diagnosis, not longer forecasts, is normally the first real win.

AI-ready asset data is the difference between a predictive maintenance programme that pays for itself and one that quietly stalls. This article sets out what AI-ready asset data actually requires — stable asset identity, a governed hierarchy, usable condition and event history, and timestamps that mean something — then gives a seven-step readiness sequence, shows where that data is really produced, and names four indicators for checking readiness before you buy another model.

Maintenance KPIs are the difference between a maintenance function that can defend its budget and one that only reports activity. This article sets out eight metrics worth tracking — PM and schedule compliance as leading indicators, MTBF and MTTR as reliability outcomes, and cost, backlog and parts availability as the commercial layer. It explains how to fix each definition so the numbers are comparable, how to set targets that survive a real week, and how to keep the underlying data honest enough that the dashboard changes decisions rather than decorating them.

The storeroom is typically the most under-managed cost centre in an asset-intensive operation — simultaneously overstocked and underprepared. This article sets out a criticality-based framework for spare parts inventory management: classify every part by the consequence of not having it and by lead time, calculate stocking levels from real work-order consumption and current supplier performance, link each part to the asset and work order it serves, and govern the catalogue with a balanced KPI set. The objective is not the smallest possible inventory, but a storeroom that can be explained at budget time and relied on during a breakdown.