Benchmarks achieved in manufacturing industry

Achievable Benchmarks

The following benchmark performances have been achieved for quality, productivity, delivery, agility, purchasing, and production control prerequisites in manufacturing industry. Comparisons are difficult since some industries are inherently better than others for good reasons, however in many cases performance can be improved significantly by following the best practices outlined on this site and the Benchmarking procedures outlined in "Introduction to Benchmarking". (Achieving these benchmarks consistently over a long period is also important.)

Links to other best practices and training at bottom of page.

Measure	Benchmark (Best we know)	Calculation	Our best project	Your Business?
Manufacturing
Quality	<25 faults per million produced	Quantity of defects x 10⁶¸ Total	<120 over a >3 month period
Productivity	666,000$ per employee pa (with significant outsourcing)	Sales ($)¸ number of employees (Will be higher if significantly automation or outsourcing / sub-contracting employed.)	>£120,000 per employee pa (with no significant automation / outsourcing)
"MOTIF" (sometimes known as "Compliance" or MRP MPS "schedule adherence")	>98% of quantity planned (work orders) actually achieved on time	Compliance to your own internal plan on-time-in-full (See Previous Best Practice 045: OTIF (on time in full)) Any work order output early / late or quantity achieved not as planned (under or over) = a miss (Implies good planning & control, and accurate provisioning for uncertainty.)	98% of all work orders 100% compliant over a >3 month period (made more difficult because scrap was occurring)
"COTIF" Delivery to customers request	100% delivered as requested (over a >6 month period)	This is based on the Customer's desired quantity / date, not the agreed / promised, quantity / date (which may be different) (See Previous Best Practice 045: OTIF (on time in full))	99.3% over a >3 month period (see below)
"POTIF" (Also Arrears / overdue sales orders are an alternative)	Zero back orders (over a >3 month period)	Measure of delivery to your Promise: Any promise inaccuracy is a miss. (In different industries a different tolerance is applied to under / over delivery and early / late delivery, which can make cross-industry comparison difficult.) (See Previous Best Practice 045: OTIF (on time in full))	99.3% (on time in full on the day, zero tolerance, over >3 month period.) (Arrears / overdue orders 0.7% of orders shipped)
"SOTIF" Planned work orders started on time	>100% (over a >3 month period)	% of planned orders started on time in full (See Previous Best Practice 045: OTIF (on time in full))	99% (on time in full on the day, zero tolerance, over >3 month period.)
Most overdue work order	<2 days	Past due work order	3 days
Demand forecast accuracy	100% time phased forecast quantity correct	Accurate quantity and date (based on hits and misses) (See Forecast Accuracy, and Demand Management). Time phased forecast quantity at lead-time vs. actual quantities ordered later.	100% time phased forecast quantity correct
Open purchase and shop order accuracy	>95%	At a stock take all orders are real and accurate counted as either accurate or inaccurate (hits & misses)	95%
Engineering data
Bill of Material (BOM)Accuracy	>98%	All single level B.O.M.s (or kit lists) are audited at build time and are either completely right or wrong (hits and misses)	>98%
Routing Data accuracy	>98%	All planned routes audited against actual and are completely right or wrong (hits & misses)	95%
Product Change	>95% on time	All product changes implemented to schedule	85%
Purchasing
Raw material availability (No shortages)	98% on time (<2% planned work order starts delayed)	Picking or Kanban shortage to production line (less than full quantity needed available) (In Kanban / line feed systems depleted buffers / unfulfilled Kanbans)	98%
Supplier on time	>98%	Full quantity on the day, with perfect quality	98%
Most overdue purchase orders	< 2 days	Full quantity, perfect quality	Not available (see note 1)
Inventory record accuracy	>95% hits	Bin quantity accurate to 2% (to system) when counted. (Within 2% is a hit. Outside 2% is a miss.)	94% hits
Flexibility
Lead-time	<2 days	Lead-time from order to delivery of complex, make to order assembly	5 days (renal dialysis machine)
Quick change / Changeover time	Average 3 minutes for very large press die	End of one batch to set up and make first satisfactory component of new batch	11 minutes for large complex former
Innovation (Ideas raised)	>10	Ideas raised per person per week. (Not all of these will make it.)	5 (over a 3 month period)
Ideas implemented	>3	Ideas implemented per week per person	2 (over a 3 month period)
Stock Turns	160 (one and a half days stock) (See note 2)	The ratio of raw material issues to total stock holding at material costs.	20 (See note 2)
		The UK Department of Trade and Industry also recommend the following further measures and they are included for completeness rather than because we believe in them:
Overall Equipment Effectiveness	Not available	Availability % x Productivity % x Quality %	We do not use this measure (See note 3)
Floor Space Utilisation	Not available	Sales revenue per square metre of floor space	We do not use this measure (See note 4)

Note 1: We tend to use raw material line shortages since this measures overall availability to where the items are needed and includes direct line feed methods and raw material stores efficiency.

Note 2: There are 3 exceptions (agricultural crops, products / processes requiring maturation, and strategic spares, the effect of which can be ring-fenced), but otherwise you should be able to achieve: an overall stock turn roughly equal to 1-2 times the process lead-time of a typical job that you are expediting through the process without queuing (E.g. If your overall manufacturing lead-time without queuing is 2 weeks, you should have about 2-4 weeks' worth of total inventory); with excellent customer service (>98% of customer orders / schedule call offs, On-Time-In-Full). For this reason it is difficult to provide outside industry comparisons for stock turn, except it is possible to quote some examples for particular industries:

Automotive component manufacturer	14 is the best we have achieved in high mix situations using MRP techniques alone (electronics assembly) (In principle an MRP1 system should be a Just In Time system.)
	160 refers to Kanban in an continuous automotive situation
FMCG's	20 is the best we have achieved in lower mix situations of assemblies using Kanban techniques.
Aerospace	16 is the best we have achieved in high mix situations using Kanban systems.

Note 3. We do not recommend the use of Overall Machine Effectiveness OME / OEE (See Previous Technique of the Week 007: CARAP (Process Effectiveness Measurement (or why OEE / OME is for the birds) because we feel it is meaningless for any resource other than the bottleneck. We also do not recommend composite measures such as this because it does not help in problem solving.

Note 4. In most of our work floor space savings may be an objective. However ongoing measurement is unlikely to influence this. These changes can only be brought about from major change programmes, which might include:

Cellular Manufacturing (See "Organisational Redesign")
Flexibility Changes (See "Agile Manufacturing")
Control systems changes such as Kanban, Participative Sales and Operations Planning etc.

__________________________________________________________

Further Reading: The following further best practice articles were mentioned in this paper:

Permanently Maintained Website Articles:

Introduction to Benchmarking

Demand Management

Organisational Redesign

Kanban

Forecast Accuracy

Participative Sales and Operations Planning

MRP

MPS

Agile Manufacturing