Operational performance

From CEOpedia
Operational performance
See also


Operational performance - is the degree to which an organisation's operations, the processes that turn inputs such as materials, labour, information and equipment into goods and services, meet the requirements placed on them by customers and by the business itself. In the operations management tradition it is usually broken down into five performance objectives: quality, speed, dependability, flexibility and cost[1].

The term sits one level below organisational performance. Profit, market share and shareholder value describe how the whole company is doing; operational performance describes how well the factory, warehouse, call centre, hospital ward or software delivery team actually works. A firm can post good profits for a while with weak operations (a strong brand or a protected market will carry it), but sooner or later late deliveries, scrap and high unit costs show up in the accounts. That is why operations managers track it so closely, and why it differs from operational excellence, which is a management philosophy and a target state rather than a measured result.

Background

For much of the twentieth century manufacturing was judged almost entirely on cost and output. Efficiency experts measured how many units a worker or a machine produced per hour, and cost accounting told managers what each unit cost. Things changed in 1969, when Wickham Skinner argued in the Harvard Business Review that manufacturing was the "missing link" in corporate strategy: top managers left the factory to technicians, while factory decisions quietly shaped what the company could and couldn't offer in the market[2]. Skinner's point was that a plant has to be designed around the way the company intends to compete, and that it can't excel at every dimension at once.

He sharpened the argument five years later in "The Focused Factory". A plant that tries to serve many markets with conflicting demands (low cost here, short lead times there, high precision somewhere else) ends up mediocre at all of them; a plant focused on a narrow set of tasks tends to outperform it[3]. This is the origin of the trade-off idea that still runs through the literature.

Japanese manufacturers in the 1970s and 1980s complicated the picture. Toyota and its suppliers seemed to deliver high quality, reliable delivery and low cost all together. Kasra Ferdows and Arnoud De Meyer studied European plants and proposed the "sand cone" model: lasting improvement comes from building capabilities in sequence, first quality, then dependability, then speed, and only then cost efficiency, each layer resting on the ones below it[4]. Trade-offs, in their view, are real at any given moment but can be shifted over time.

By the 1990s the debate had moved from "what to measure" to "how to design a measurement system". Andy Neely, Mike Gregory and Ken Platts reviewed the field and described performance measurement as quantifying both the efficiency and the effectiveness of action, warning that systems built only from financial ratios miss most of what operations do[5].

The five performance objectives

Nigel Slack, Stuart Chambers and Robert Johnston organise operational performance around five objectives. Each has an external effect, felt by customers, and an internal effect, felt inside the operation[1].

Objective Meaning External effect Internal effect
Quality Doing things right, conforming to specification Satisfied customers, fewer complaints and returns Less rework and scrap, more dependable processes
Speed Doing things fast, short time between request and delivery Shorter lead times for customers Less work-in-progress and inventory, lower risk
Dependability Doing things on time, keeping promises Reliable delivery dates Fewer disruptions, more stable schedules, saves time and money
Flexibility Being able to change what is done, how, or when New products, wider range, volume and delivery changes Faster response to breakdowns and shortages
Cost Doing things cheaply Lower prices or higher margins Productive use of resources

The internal effects matter more than they first seem. Good quality reduces cost because nothing has to be made twice. Dependability saves time, because a supervisor who isn't chasing late batches can plan. Speed reduces inventory, and inventory ties up cash. Cost, in this sense, is partly the result of the other four.

Not all objectives weigh the same for every operation. Terry Hill's distinction between "order winners" and "order qualifiers" is often used here: qualifiers are the minimum level a firm must reach to be considered at all, while winners are what actually tips the customer's decision. For a budget airline, punctuality may be a qualifier and price the winner. For an emergency spare-parts supplier, speed wins and price barely matters. The point links operational performance directly to operations strategy.

Common measures

Each objective is broad, so it gets translated into specific, measurable performance indicators. Some of the most widely used:

  • Overall equipment effectiveness (OEE) - the share of planned production time that is truly productive, combining availability, performance (speed) and quality. Developed within total productive maintenance by Seiichi Nakajima[6].
  • Throughput - output per unit of time (units per hour, orders per day, patients per shift). In the theory of constraints throughput is set by the bottleneck, not by the average machine.
  • Cycle time - the time to complete one unit or one process step; compared with takt time it shows whether the process can keep pace with demand.
  • Lead time - the elapsed time from order to delivery, the customer's view of speed.
  • On-time delivery (OTD) - percentage of orders delivered on or before the promised date; the standard dependability measure.
  • First pass yield (FPY) - share of units that pass all quality checks first time, without rework.
  • Defect rate (often in parts per million) and the cost of poor quality.
  • Unit cost - total operating cost divided by good output.
  • Productivity - output divided by input, such as units per labour-hour or, more broadly, total factor productivity.
  • Changeover time and product range - typical flexibility measures.

None of these is perfect on its own. On-time delivery can be pushed up by quoting long lead times. Throughput can rise while quality falls. That's why the measures are usually read together.

Formula of overall equipment effectiveness

OEE is probably the single most cited operational measure in manufacturing. It multiplies three ratios[6]:

\[ OEE = Availability \times Performance \times Quality \]

where

\[ Availability = \frac{Run\ time}{Planned\ production\ time} \]

\[ Performance = \frac{Ideal\ cycle\ time \times Total\ count}{Run\ time} \]

\[ Quality = \frac{Good\ count}{Total\ count} \]

After cancelling terms, the product simplifies to

\[ OEE = \frac{Good\ count \times Ideal\ cycle\ time}{Planned\ production\ time} \]

which reads as: what fraction of the planned time was spent making good parts at the ideal rate. Nakajima linked the three ratios to six big losses: breakdowns and setup/adjustment reduce availability; idling, minor stops and reduced speed reduce performance; defects and start-up losses reduce quality. A value of about 85% is widely quoted by practitioners as "world class", though the figure depends heavily on how the inputs are defined.

Example of operational performance measurement

Hypothetical example. A packaging line runs one 8-hour shift (480 minutes) with a planned 30-minute break. Planned production time is therefore 450 minutes. During the shift the line loses 30 minutes to a breakdown and 15 minutes to a changeover, so run time is 405 minutes. The ideal cycle time is 0.5 minutes per unit (120 units per hour). The line produces 720 units, of which 36 are rejected.

  • Availability = 405 / 450 = 90.0%
  • Performance = (0.5 × 720) / 405 = 360 / 405 = 88.9%
  • Quality = 684 / 720 = 95.0%
  • OEE = 0.900 × 0.889 × 0.950 = 76.0%

Check via the short formula: (684 × 0.5) / 450 = 342 / 450 = 76.0%. Same answer.

The plant manager also keeps a small KPI dashboard. Conversion cost for the shift (labour, energy, maintenance, overhead) is €6,840; six operators work 7.5 productive hours each, i.e. 45 labour-hours. Last month the plant shipped 1,840 of 2,000 orders on time. After a maintenance and setup-reduction project, the line loses only 30 minutes per shift, runs at 90% of ideal speed and produces 756 units with 30 rejects.

KPI Before After
Availability 90.0% 93.3%
Performance 88.9% 90.0%
Quality (first pass yield) 95.0% 96.0%
OEE 76.0% 80.7%
Good units per shift (throughput) 684 726
Labour productivity (good units per labour-hour) 15.2 16.1
Conversion cost per good unit €10.00 €9.42
On-time delivery (last month) 92.0% -

Unit cost drops by about 5.8% without a single euro of extra spending, simply because the same fixed shift cost is spread over 42 more good units. If each extra unit carries a contribution of €4 and the line runs 500 shifts a year, the gain is 42 × 4 × 500 = €84,000 per year. Small percentages. Real money.

Link to financial results

Operational measures are leading indicators; financial ones lag. Better quality cuts scrap, rework and warranty claims, so it lowers cost of goods sold. Shorter lead times and less work-in-progress shrink inventory and free working capital. Higher OEE lets a plant meet extra demand without buying another machine, which improves asset turnover and return on investment. Dependability keeps customers, and keeping customers is cheaper than winning new ones.

The chain isn't automatic, though. Extra capacity only becomes profit if there is demand to fill it; otherwise the line just finishes its work earlier, or worse, builds inventory nobody ordered. This is a standard warning in just-in-time and lean thinking: producing more than the market takes is waste, not performance.

Frameworks for managing operational performance

  • Balanced scorecard - Robert Kaplan and David Norton added an internal business perspective next to the financial, customer, and innovation and learning perspectives, asking managers to identify the internal processes the firm must excel at and measure them alongside financial results[7]. Cycle time, quality and productivity are typical measures in that perspective.
  • Lean management - focuses on flow and waste removal; uses lead time, inventory, takt time and value-added ratio, and relies on kaizen for continuous small improvements.
  • Six Sigma - targets variation and defects through the DMAIC cycle and statistical process control; its home metrics are defects per million opportunities and process capability.
  • Benchmarking - compares a plant's measures with competitors, sister plants or best-in-class operations in other industries, so that a 76% OEE or 92% on-time delivery means something beyond its own history.
  • Performance measurement system design - the approach of Neely and colleagues, who stress that measures should be derived from strategy, kept few in number and reviewed as strategy changes[5]. The wider topic is covered in measurement of performance.

Advantages of measuring operational performance

  • Makes problems visible early, before they reach the income statement.
  • Gives a shared language between shop floor, finance and top management.
  • Shows where improvement money pays off; OEE losses, for instance, point straight at breakdowns, speed or defects.
  • Supports customer satisfaction by tracking what customers actually feel: delivery, quality, lead time.
  • Allows fair comparison between sites, shifts and periods, and supports benchmarking.
  • Links daily work to strategy when the measures are chosen from the objectives that win orders.

Limitations of operational performance measurement

  • Trade-offs - as Skinner warned, pushing one objective can hurt another[3]. Longer production runs lower unit cost but cut flexibility and raise inventory. The sand cone model suggests these can be softened over time, not abolished[4].
  • Gaming - when a number becomes a target, people learn to move the number. Ideal cycle times get set conservatively so that performance looks high; downtime gets reclassified as "planned"; delivery dates get renegotiated so that OTD stays green.
  • Local optimisation - a department that maximises its own utilisation may starve or flood the next one. High OEE on a non-bottleneck machine adds nothing to plant throughput, a point made forcefully by the theory of constraints.
  • Definitional differences - two plants reporting OEE may use different rules for breaks, changeovers or minor stops, which makes benchmarking misleading[6].
  • Measurement overload - dashboards with fifty indicators dilute attention. Nobody acts on all of them.
  • Short-term bias - operational measures reward this month's output; investment in training, maintenance or new processes may depress them before it pays off.
  • Service operations - quality and flexibility in services are harder to quantify than in a factory, and customer perception doesn't always follow the internal numbers.

Related approaches

Operational performance connects to several neighbouring ideas. Efficiency covers the input-output side, while effectiveness asks whether the right outputs are produced at all. Quality management supplies most of the tools for the quality objective. Supply chain management extends the five objectives beyond the factory gate to suppliers and distributors, where dependability and speed often depend on partners. And mass customization is, in a sense, an attempt to win on flexibility without losing on cost, which is exactly the kind of trade-off shift Ferdows and De Meyer described.

References

Footnotes

  1. ↑ Slack N., Chambers S., Johnston R., Operations Management, 6th edition, Financial Times Prentice Hall, 2010
  2. ↑ Skinner W., Manufacturing - Missing Link in Corporate Strategy, Harvard Business Review, 1969
  3. ↑ Skinner W., The Focused Factory, Harvard Business Review, 1974
  4. ↑ Ferdows K., De Meyer A., Lasting improvements in manufacturing performance: In search of a new theory, Journal of Operations Management, 1990, pp. 168-184
  5. ↑ Neely A., Gregory M., Platts K., Performance measurement system design, International Journal of Operations & Production Management, 1995, pp. 80-116
  6. ↑ Muchiri P., Pintelon L., Performance measurement using overall equipment effectiveness (OEE), International Journal of Production Research, 2008, pp. 3517-3535
  7. ↑ Kaplan R.S., Norton D.P., The Balanced Scorecard - Measures that Drive Performance, Harvard Business Review, 1992
Operational performance — recommended articles
Operational excellence — Organisational performance — Measurement of performance — Total productive maintenance — Balanced scorecard — Lean management — Six Sigma — Benchmarking
Retrieved from CEOpedia