Supplier Delivery and Quality Metrics

Why it matters

Whether a supplier delivers well depends on what is counted as success. Even if the vehicle arrives on the promised day, the production plan can still be disrupted when part of the material is missing. Material that arrives in full but late has also not met the need on time. For this reason the on-time and in-full delivery measure looks for both conditions in the same order. Placing time and quantity ratios that each look high side by side does not show joint success. The basic question of the delivery measure is whether the order reached the defined delivery point by its due date and in the full quantity. For usability, quality acceptance is verified separately. The record that answers this question needs a different denominator from the number of vehicles or the total number of units received. The measure design should explain this distinction before any calculation.

Quality is a separate conformance condition. If part of material that arrived on time and in full is defective, the delivery measure and the quality measure can give different results. This is not a contradiction; the two measures are measuring different conditions. But if the results are to be combined, time, quantity and quality are examined together in the same order record. Directly multiplying a unit-based defect-free rate by an order-based delivery success generally does not give the correct joint rate. Defects may be concentrated within successful deliveries. For joint success, each order's conditions are evaluated separately. In this way, in the discussion with the supplier, instead of a general verdict of good or bad, it is shown which condition was broken and which orders were affected.

The choice of denominator changes the result considerably. An order may have arrived with several vehicles; counting vehicle movements as separate chances of success gives more weight in the measurement to orders that are split into many parts. In the same way, multiplying the number of easy orders with small quantities can keep a total-unit problem from showing. Order-based success and a unit-based fulfillment rate can be presented side by side; but one does not replace the other. The number of orders whose delivery obligation fell due is fixed at the start of the period. An order that did not arrive stays in the denominator. Examining only the shipments that arrived can mean removing the most serious delays from the report. Missing records must not be silently accepted as success.

The purpose of measurement is to make a shared problem visible before it is to produce a penalty. The requested date, the confirmed date and the later changed date are kept separate. Moving the date of a late order forward retroactively does not correct past performance. Late changes made by the buyer are not recorded as supplier faults; the cause class is kept separately. The receiving window, the condition for early delivery, the quantity tolerance and the time the quality decision closes are explained from the start. The success given in this lesson is calculated by the defined rule; no universal acceptance threshold is proposed for every business. The result should be read together with the detailed records that will set the development priority. The meaning of a number is only as reliable as the definition behind it.

Formula