Consultancy Report Example: Improving Operational Efficiency at a UK Manufacturing Firm

Executive Summary

Pennine Precision Components Ltd (PPC) engaged the consulting team to diagnose the causes of its deteriorating delivery performance and eroding margins, and to recommend a practical, costed improvement programme. Over a four-week engagement, the team combined structured interviews, direct observation, value stream mapping and the analysis of twelve months of production data to build an evidence-based picture of the current state of operations.

The diagnosis identified the computer numerical control (CNC) machining cell as the binding constraint on the entire value stream. Overall equipment effectiveness (OEE) at the cell stood at 52 per cent, well below the 85 per cent benchmark widely regarded as world-class, with availability losses the single largest contributor. Average changeover times of 95 minutes, an internal scrap rate of 6.8 per cent and reactive, run-to-failure maintenance practices together explained the majority of the lost capacity. These losses propagated downstream, producing an on-time-in-full (OTIF) delivery rate of 78 per cent and a manufacturing lead time of 21 working days against a customer expectation of 12.

The report recommends a phased programme built on single-minute exchange of die (SMED), total productive maintenance, error-proofing and pull-based scheduling, prioritised by cost and payback. Implemented in full, the programme is projected to raise cell OEE above 70 per cent and OTIF above 95 per cent within twelve months, at modest capital cost.

Introduction

Background to the Engagement

Pennine Precision Components Ltd is a privately owned, tier-two manufacturer of precision-machined components supplying the UK automotive and aerospace sectors. Founded in 1994 and based in West Yorkshire, the firm employs approximately 180 staff and reported turnover of £24 million in its most recent financial year. Its competitive position rests on an ability to deliver complex, low-to-medium-volume parts to tight tolerances and at short notice, rather than on price alone.

Over the preceding eighteen months, senior management had observed a steady decline in commercial performance. Gross margin had fallen by roughly four percentage points, two significant customers had placed their contracts “on review” while citing repeated late deliveries, and the volume of formal quality complaints had risen. Concerned that these symptoms shared a common operational cause, the board commissioned this consultancy engagement to establish the root causes of the decline and to set out a prioritised plan for recovery.

Purpose and Scope

The purpose of this report is to present the findings of that engagement and to translate them into actionable, costed recommendations. The scope was agreed with the sponsor as the internal manufacturing value stream, from the receipt of raw bar stock through to the despatch of finished components. It deliberately excluded the firm’s supplier base and its commercial pricing strategy, both of which were the subject of separate work streams. The engagement ran for four weeks and was sponsored by the Operations Director, who provided access to data, staff and the shop floor.

The report is structured as follows. The methodology section explains the mixed-methods approach used to gather and triangulate the evidence. The findings section presents the quantitative and qualitative results, concentrating on the machining cell identified as the critical constraint. The analysis section interprets these findings through established operations-management theory, and the recommendations section sets out a costed, phased improvement programme. A short conclusion then reflects on the limitations of the work and the conditions required for its success.

Methodology and Approach

The engagement adopted a mixed-methods diagnostic approach, combining qualitative and quantitative evidence to reduce the risk of drawing conclusions from any single, potentially biased source. This practice of triangulation is widely recommended in operations improvement, where shop-floor perceptions and recorded data frequently diverge (Slack, Brandon-Jones and Johnston, 2019). Four complementary methods were used.

• Document and data review. Twelve months of production records were extracted from the firm’s enterprise resource planning (ERP) system, covering output volumes, scrap and rework, machine downtime logs, changeover records and despatch performance. These data established the quantitative baseline against which improvement could later be measured.
• Direct observation (Gemba walks). The team spent eight days on the shop floor observing the flow of work, timing a sample of changeovers and recording interruptions to production. Observation is essential because recorded downtime codes often understate minor stoppages and small speed losses, which accumulate into significant lost capacity (Nakajima, 1988).
• Semi-structured interviews. Fourteen staff were interviewed, spanning operators, team leaders, maintenance technicians, production planners and senior managers. These conversations surfaced tacit knowledge and competing explanations for poor performance that the data alone could not reveal.
• Value stream mapping. A current-state map was constructed for the highest-volume product family, quantifying process times, inventory levels and the ratio of value-adding time to total lead time (Rother and Shook, 1999).

Overall equipment effectiveness was selected as the principal performance metric for the constraint resource because it decomposes losses into three intuitive components and therefore directs attention to the specific nature of each loss, rather than reporting an undifferentiated utilisation figure (Nakajima, 1988). It is calculated as the product of the three components:

OEE = Availability × Performance × Quality

Here, availability is the ratio of actual running time to planned production time; performance compares actual output with the theoretical maximum at the machine’s rated speed; and quality is the proportion of output that meets specification at the first attempt. Expressing each loss separately allowed the team to identify which component offered the greatest opportunity for recovery.

Two limitations of the methodology should be acknowledged at the outset. First, the downtime data held in the ERP system relied on manual operator entry and was incomplete for minor stoppages; the observational study was used to correct for this where possible. Second, the four-week window captured a representative but not exhaustive production mix, so seasonal variation could not be fully assessed. These constraints are revisited in the conclusion.

Findings

The Machining Cell as the Critical Resource

Both the value stream map and the production data pointed unambiguously to the CNC machining cell as the operation’s bottleneck. It was the only resource consistently running at or near its available capacity, it held the largest queue of work waiting to be processed, and every major product family routed through it. In the language of the theory of constraints, it was the “drum” that set the pace for the entire plant (Goldratt and Cox, 2004). The detailed findings therefore concentrate on this cell, since improvement elsewhere would yield little benefit while the constraint remained unaddressed.

Overall Equipment Effectiveness at the Cell

A two-week OEE study of the cell’s four machining centres produced the results summarised in Table 1. The aggregate OEE of 52 per cent sits far below the 85 per cent figure conventionally regarded as world-class, and implies that almost half of the cell’s paid, planned capacity was being lost to the three categories of loss.

Table 1: OEE breakdown for the machining cell (illustrative figures).

Availability was the weakest of the three components and therefore represented the largest single opportunity. The downtime logs, once corrected by the observational study, attributed availability loss principally to two sources: lengthy changeovers between product variants, and unplanned machine breakdowns arising from a reactive maintenance regime.

Changeover and Set-up Losses

The cell averaged 3.2 changeovers per machine per day, each taking a mean of 95 minutes. Observation of a sample of changeovers found that a substantial proportion of this time was consumed by “internal” work – tasks performed while the machine was stopped that could, with adequate preparation, have been carried out while it was still running. Tooling and fixtures were frequently located and retrieved during the changeover itself rather than staged in advance, and the first component produced after each changeover typically required several adjustment cycles before it met specification. Changeovers were treated as an unavoidable craft activity rather than a process that could be studied and improved.

Quality and Scrap Performance

First-time quality at the cell was 89 per cent, giving an internal scrap and rework rate of 6.8 per cent of output. A Pareto analysis of twelve months of non-conformance records showed that three defect modes accounted for roughly three-quarters of all scrap, and that a disproportionate share of defects occurred on the first parts produced immediately after a changeover. This finding linked the quality problem directly to the set-up problem: unstable changeovers were generating not only lost time but also defective parts, each of which consumed further constraint capacity in rework or replacement.

Material Flow and Work-in-Progress

The current-state value stream map revealed a manufacturing lead time of 21 working days for the reference product family, of which only around 1.5 days represented genuine value-adding processing time – a value-added ratio below eight per cent. The remainder was almost entirely queue time, with approximately £1.9 million of work-in-progress inventory standing idle between processes, equivalent to roughly 34 days of stock. Large batch sizes, themselves a rational response to long changeover times, inflated these queues and lengthened lead times still further.

Delivery Performance and Its Commercial Impact

The internal conditions described above translated directly into the symptom that customers experienced: unreliable delivery. OTIF performance stood at 78 per cent, meaning that almost one order in five arrived late or incomplete. Because PPC competes on responsiveness rather than price, this shortfall struck at the heart of its value proposition and explains why two customers had placed their contracts under review. Interviews with the planning team confirmed that late orders were routinely “expedited” through the plant, disrupting the schedule, displacing other work and generating yet more lateness – a firefighting pattern that consumed management attention without addressing the underlying constraint. The commercial cost of poor operations was therefore considerably greater than the direct cost of scrap and downtime alone, since it also placed future revenue at risk.

Analysis and Discussion

Interpreting the Losses Through Lean Theory

Viewed through the lens of lean thinking, the findings describe an operation in which a large share of activity adds cost without adding value. The seven classic wastes identified by Ohno (1988) and popularised by Womack and Jones (2003) were all visible at PPC, as summarised in Table 2. Mapping the evidence onto this framework is useful because it reframes the firm’s problems not as isolated failings but as predictable symptoms of a system that has not been designed for flow.

Table 2: The seven wastes observed at PPC.

The Constraint and Its Implications

The theory of constraints offers a powerful principle for prioritising the improvement effort: an hour lost at the bottleneck is an hour lost for the entire system and can never be recovered, whereas an hour saved at a non-constraint is largely an illusion (Goldratt and Cox, 2004). Because the machining cell governs the throughput of the whole plant, interventions that recover its lost availability will increase saleable output, shorten lead times and improve delivery reliability simultaneously. Conversely, investment in faster upstream processes or additional non-constraint equipment would simply enlarge the queue in front of the cell. This logic concentrates the recommendations on the cell and on subordinating the rest of the operation to it.

Root Causes of Availability Loss

Applying iterative questioning of the “five whys” type to the availability losses revealed two distinct but related root causes. Changeovers were lengthy not because the work was inherently complex but because internal and external set-up activities had never been separated and standardised; a disciplined application of the SMED method, which converts internal tasks into external ones wherever possible, routinely halves changeover times in comparable operations (Shingo, 1985). Breakdowns, meanwhile, were frequent because maintenance was essentially reactive: equipment was run until it failed, and operators had no role in its upkeep. Total productive maintenance addresses precisely this pattern by combining planned preventive maintenance with autonomous maintenance performed by operators, thereby stabilising availability at source (Nakajima, 1988; Liker, 2004).

The Batch-Size Trap and the Cost of Poor Quality

The analysis also exposed a self-reinforcing cycle that explains why the firm’s problems had proved so stubborn. Long changeovers made small batches uneconomic, so planners ran large batches; large batches inflated work-in-progress and lead times, and increased the quantity of material exposed to any process error before it was detected. At the same time, each changeover carried a high risk of first-off defects, which discouraged frequent changeovers still further. The result was a system locked into long lead times, high inventory and poor responsiveness. The decisive insight is that reducing changeover time breaks this cycle at its root: once set-ups are quick, batches can be reduced without sacrificing capacity, which simultaneously lowers inventory, shortens lead time and – by cutting the number of parts exposed to each error – reduces the cost of poor quality (Shingo, 1985; Womack and Jones, 2003; Bicheno and Holweg, 2016).

Why Earlier Improvement Efforts Had Not Taken Hold

Interviews revealed that PPC had attempted improvement before: a 5S initiative two years earlier and a short-lived “lean week” had both faded within months. Understanding why matters, because it shapes how the present programme should be designed. Three organisational factors emerged. First, previous efforts had targeted areas that were visible rather than the areas that constrained throughput, so even successful changes produced little measurable benefit and lost support. Second, improvements were run as detached events rather than being sustained through standard work and daily management, so new habits decayed once the facilitators left. Third, operators had been the objects of change rather than its agents; the knowledge needed to shorten changeovers and prevent breakdowns sat with the people doing the work, yet they had not been engaged in designing the solutions. The literature on lean transformation consistently identifies this gap between tool application and genuine cultural and managerial change as the principal reason improvement programmes stall (Liker, 2004; Hines, Holweg and Rich, 2004). Any recommendation must therefore address the management system, not merely the technical methods.

Quantifying the Prize

The scale of the opportunity can be illustrated by a simple bridge. If a SMED programme halves changeover time and a preventive maintenance regime reduces unplanned breakdowns, availability could plausibly rise from 71 per cent to around 85 per cent. Combined with modest gains in performance (to roughly 88 per cent, through fewer minor stoppages) and in quality (to roughly 95 per cent, through error-proofing of first-off parts), the resulting OEE would be approximately 0.85 × 0.88 × 0.95, or around 71 per cent. This would represent an increase of nineteen percentage points in effective capacity at the constraint, achieved largely through better use of existing assets rather than capital investment. Such an interpretation is consistent with the wider evidence that the largest returns from lean transformation come from improving flow and reducing variability rather than from adding resource (Hines, Holweg and Rich, 2004).

Recommendations

The recommendations below form a coherent, phased programme rather than a menu of independent initiatives. They are sequenced so that early, low-cost interventions stabilise the constraint and build momentum, while later changes to scheduling and inventory lock in the gains. Each recommendation is mapped to the loss it addresses, and the full set is prioritised in Table 3.

1. Launch a SMED programme at the machining cell. Video a representative changeover, separate internal from external set-up, convert internal tasks to external wherever possible, and standardise the remaining steps with a pre-staged tooling trolley. Target a 50 per cent reduction in changeover time within three months.
2. Introduce total productive maintenance. Establish a planned preventive maintenance schedule for the four machining centres and train operators in autonomous maintenance (daily cleaning, inspection and lubrication) so that emerging faults are detected before they cause breakdowns.
3. Implement 5S and visual management. Organise the cell so that tools, fixtures and materials have a defined location, supporting both faster changeovers and more stable, repeatable work.
4. Error-proof first-off quality. Introduce standard work and simple poka-yoke checks for the first parts after each changeover, targeting the three defect modes that account for most scrap, so that quality losses fall as changeover frequency rises.
5. Reduce batch sizes and move toward pull scheduling. Once changeovers are short, progressively reduce batch sizes and introduce a pull-based signal (such as a kanban or constant-work-in-progress limit) to cap inventory and shorten lead time.
6. Establish daily performance management. Display OEE, scrap and delivery performance on a visual board at the cell and hold a short daily review, supported by a structured continuous-improvement (kaizen) routine, to sustain the gains and surface new problems quickly.

Table 3: Prioritised recommendations with indicative cost and payback (illustrative).

Taken together, and on the illustrative figures above, the programme is projected to raise cell OEE from 52 to above 70 per cent, lift OTIF from 78 to above 95 per cent, reduce manufacturing lead time from 21 to approximately 12 working days, cut work-in-progress by around 40 per cent and reduce the scrap rate from 6.8 to below 3 per cent. These outcomes would directly address the delivery failures and margin erosion that prompted the engagement, at a total indicative capital cost well below £100,000.

Governance and Implementation Roadmap

Given that earlier efforts had faded, the programme should be governed as a sustained transformation rather than a sequence of isolated events. It is recommended that a cross-functional improvement team be formed, led by a respected cell team leader, sponsored visibly by the Operations Director and meeting weekly to review progress against a small number of agreed measures. The work falls naturally into three phases. In the first quarter, the team should stabilise the constraint through SMED, 5S and first-off error-proofing, since these deliver rapid, visible returns that build confidence. In the second and third quarters, it should embed total productive maintenance and begin reducing batch sizes. From the third quarter onward, it should introduce pull scheduling and consolidate the daily management routine so that improvement becomes continuous. Crucially, operators should be trained and involved throughout, so that ownership of the new methods rests with those who use them.

Three principal risks should be managed actively. The first is that short-term demand pressure diverts the team back into firefighting; a visible sponsor and protected improvement time mitigate this. The second is that gains are not sustained once attention moves on, which standard work and the daily review are designed to prevent. The third is that measurement remains unreliable; installing automated data capture at the cell is therefore an early priority, so that decisions rest on evidence rather than impression.

Conclusion

This report has diagnosed the operational causes of PPC’s declining delivery performance and margins, and has set out a prioritised programme to address them. The central conclusion is that the firm’s problems are not a collection of unrelated faults but the predictable consequence of a single under-performing constraint. The CNC machining cell, operating at an OEE of 52 per cent, loses almost half of its planned capacity to lengthy changeovers, reactive maintenance and unstable quality, and these losses cascade into long lead times, excess inventory and unreliable delivery. Because the cell governs the throughput of the whole plant, a focused programme of SMED, total productive maintenance, error-proofing and pull scheduling offers substantial improvement at modest cost.

Several conditions will determine whether these benefits are realised. Sustained leadership commitment and a visible sponsor are essential, as is genuine engagement of operators, who hold much of the knowledge needed to shorten changeovers and prevent breakdowns. The gains must also be embedded through standard work and daily management rather than treated as a one-off project, since improvements that are not sustained tend to decay. Finally, the limitations noted earlier – incomplete minor-stoppage data and a single-season observation window – mean that the OEE baseline should be re-measured once reliable, automated data capture is in place, so that the programme’s impact can be verified against a firm foundation.

References

• Bicheno, J. and Holweg, M. (2016) The Lean Toolbox: A Handbook for Lean Transformation. 5th edn. Buckingham: PICSIE Books.
• Goldratt, E.M. and Cox, J. (2004) The Goal: A Process of Ongoing Improvement. 3rd edn. Great Barrington, MA: North River Press.
• Hines, P., Holweg, M. and Rich, N. (2004) ‘Learning to evolve: a review of contemporary lean thinking’, International Journal of Operations & Production Management, 24(10), pp. 994–1011.
• Liker, J.K. (2004) The Toyota Way: 14 Management Principles from the World’s Greatest Manufacturer. New York: McGraw-Hill.
• Nakajima, S. (1988) Introduction to TPM: Total Productive Maintenance. Cambridge, MA: Productivity Press.
• Ohno, T. (1988) Toyota Production System: Beyond Large-Scale Production. Portland, OR: Productivity Press.
• Rother, M. and Shook, J. (1999) Learning to See: Value Stream Mapping to Add Value and Eliminate Muda. Brookline, MA: Lean Enterprise Institute.
• Shingo, S. (1985) A Revolution in Manufacturing: The SMED System. Cambridge, MA: Productivity Press.
• Slack, N., Brandon-Jones, A. and Johnston, R. (2019) Operations Management. 9th edn. Harlow: Pearson.
• Womack, J.P. and Jones, D.T. (2003) Lean Thinking: Banish Waste and Create Wealth in Your Corporation. 2nd edn. London: Simon & Schuster.