Orlicky's MRP turned a production plan into timed component orders

Before MRP, factories used reorder-point and economic-order-quantity rules, which treat each item on its own. But a product is made of many parts with a fixed structure, so reordering each part independently let finished goods and components drift into glut and shortage.

Joseph Orlicky, an IBM engineer, set out the principles of material requirements planning in 1964. MRP reads the master production schedule, works through the bill of materials, offsets lead times, and works out exactly when each part must be ordered.

Black & Decker was the first manufacturing user, with Dick Alban leading the project. By 1975 MRP was in about 700 companies and nearly 8,000 by 1981; Oliver Wight later extended it into MRP II, adding capacity, master scheduling and financial planning.

The payoff was a computer computing requirements that no manual planner could, cutting inventory while protecting delivery. The generalisable idea: derive lower-level demand from the top-level plan rather than from each part's own history.

Derive what's needed from the plan, not from part history

Part demand is dependent — it comes from what is built — so an independent reorder point is the wrong trigger.

Lead-time offset places orders just in time, so stock is low but never missing.

It turns a finished-goods plan into an automatic purchasing and shop-floor action list.

Computerised MRP enables rapid recalculations that a manual planner simply could not do.

Components have dependent demand: order them from the finished-goods plan with lead-time offsets, not from each part's own reorder point. That is how you cut stock without missing deliveries.

MRP spawned MRP II and ERP and became the default manufacturing planning method. By 1989 about a third of the software industry was MRP II software worth roughly $1.2 billion, and firms that fully adopted it often reported returns on investment near 200%.