Before you Buy Another Machine: Find Hidden Capacity in Your Factory

Understanding your current production capacity. Easy right?

For many manufacturers, the answer is surprisingly difficult to establish.

A machine might appear busy. Operators might feel stretched. Production schedules might be full. Yet the factory can still be losing significant amounts of productive time through short stops, waiting, changeovers, material shortages, inspection, operator constraints and bottlenecks elsewhere in the process.

The challenge is separating capacity that exists in theory from capacity that can realistically be recovered, and as our head of finance (who previously worked in manufacturing) frequently reminds us, that distinction matters.

Capacity is becoming a more expensive problem to solve

Manufacturing productivity is fundamentally about generating more output from the resources already available. At its simplest:

Productivity = Output ÷ Input

Those inputs might include labour, machinery, energy, materials and capital.

The principle is straightforward. In practice, improving productivity is much harder because the constraint is rarely obvious, this matters particularly when manufacturers are making decisions about growth.

Increasing capacity through additional machinery can involve hundreds of thousands of pounds of investment, long lead times, installation, training and additional staffing. Adding another shift can increase labour costs significantly, whilst overtime might solve an immediate problem, but it rarely represents a sustainable capacity strategy.

Before taking any of these routes, manufacturers need to understand whether their existing production system is genuinely at capacity. That means looking beyond how much equipment you own and understanding how effectively that equipment is actually being used.

Installed capacity and usable capacity are not the same thing

Imagine a CNC machine is available for two eight-hour shifts each weekday.

On paper, that gives roughly 160 hours of available production time over a two-week period. But those 160 hours are not necessarily 160 productive hours.

There might be:

• 18 hours of planned maintenance and breaks

• 15 hours of changeovers and setup

• 12 hours waiting for material

• 10 hours waiting for an operator

• 8 hours of breakdowns

• 6 hours of inspection delays

• 5 hours where no job was scheduled

The machine could therefore deliver substantially fewer productive hours than its theoretical capacity suggests. This is where capacity discussions can become misleading.

A machine running 50% of the time does not automatically mean the business can double its output.  Some of the remaining 50% will be necessary or unavoidable.

Therefore, the useful question is not ‘How much unused capacity do we have’? It’s actually ‘How much recoverable capacity do we have’?

Recoverable capacity is the production time currently being lost to constraints that could realistically be reduced or removed. That is a much more useful number.

Low utilisation does not always mean the machine is the problem

One of the dangers of looking at individual machine utilisation is that it can encourage manufacturers to optimise the wrong part of the process.

A machine might only be productive for 40% of a shift because:

• Material is not arriving quickly enough;

• An operator is responsible for several machines;

• Inspection is creating a queue;

• Jobs are not being released from planning;

• Changeovers are taking longer than expected;

• Another process upstream is restricting work;

• Finished components cannot move downstream quickly enough.

In each case, the machine is recording lost production time, but buying a faster machine would not solve the problem.

This is why production capacity should be viewed as a system, rather than as a collection of individual assets as the output of the factory is determined by the interaction between people, machines, materials and processes. And ultimately, it is determined by the constraint.

Find the constraint before trying to increase capacity

This principle sits at the heart of the Theory of Constraints.

The theory argues that the performance of a production system is limited by a relatively small number of constraints — often one dominant constraint at any given point, for example, lets imagine a production process with four stages:

Increasing machining capacity from 70 to 100 parts per hour might sound like an improvement, but factory throughput is still restricted by inspection at 50 parts per hour.

In that situation, improving the machining process may increase local utilisation without increasing overall output. The same principle applies across much more complicated factories. A bottleneck may be a machine, but it may equally be:

• an operator;

• a crane;

• an inspection process;

• a changeover;

• material availability;

• a planning decision;

• an internal logistics process.

Until that constraint is understood, investment decisions are partly based on assumptions.

Measure lost production in hours, not just percentages

Manufacturing teams are surrounded by percentages. Utilisation. OEE. Availability. Performance. Efficiency. These are all useful measures, but percentages can sometimes make an operational problem feel abstract. Production hours are often easier to act on, as illustrated in these two statements:

Machine utilisation is 46%. Versus: This production area lost 310 potentially productive machine hours last month.

The second immediately raises practical questions.

• Where did those hours go?

• How many were unavoidable?

• Which losses were repeated?

• What would happen if 50 hours were recovered?

• What additional output could those hours create?

• Could overtime be reduced?

• Would an additional shift still be necessary?

• Could the business accept more orders?

• Could delivery performance improve?

Turning percentages into time makes the improvement opportunity feel much more tangible.

Not every lost hour is worth recovering

There is another important distinction, the goal should not be to remove every minute of downtime. That is unrealistic and, in many cases, undesirable.

• Machines need maintenance.

• Operators need breaks.

• Products need inspection.

• Changeovers are necessary where production is varied.

Some equipment may deliberately have spare capacity because it protects the overall production system.

The objective is therefore not maximum machine utilisation. It is the right utilisation in the right part of the factory.

A machine running at 100% utilisation can actually create problems if the next process cannot absorb the output. Work-in-progress builds up, lead times increase, and operators become busy producing inventory rather than meeting demand.

Capacity improvement should therefore focus on losses that affect throughput, customer delivery or cost.

A practical way to calculate recoverable capacity

A useful starting point is to break production time into four categories.

1. Available time

The total time the machine or production area could theoretically operate.

For example: 2 shifts × 8 hours × 5 days = 80 hours per week

2. Planned non-production time

Known periods when production is not expected. This might include:

• Planned maintenance;

• Breaks;

• Planned cleaning;

• Training;

• Agreed shut down periods.

If this totals 10 hours, it would work out as: 80 available hours - 10 planned hours = 70 scheduled production hours

3. Productive time

The amount of scheduled time where the machine is actually producing.

If the machine produces for 45 hours: 45 ÷ 70 = 64% utilisation of scheduled production time

4. Lost productive time

The remaining 25 hours should then be categorised. For example:

That does not mean all 25 hours can be recovered.

But it does give the production team somewhere useful to start. If material waiting can realistically be reduced by four hours, changeovers by two and operator delays by three, the factory has identified:

9 hours of realistically recoverable capacity each week.

Across 50 working weeks, that represents roughly: 450 additional production hours per year.

That is a very different conversation from simply saying utilisation is 64%.

Real manufacturing examples: Finding capacity before adding it

We see the difference this can make when manufacturers begin measuring what actually happens on the factory floor rather than relying purely on scheduled production time.

Identifying more than 400 hours of potential capacity

Our customer, HadFab, a UK steel fabricator, began monitoring 13 CNC machines to establish a reliable picture of production performance.

The data helped identify a significant bottleneck as well as recurring losses caused by activities such as inspection, loading, unloading and waiting for crane movements. Importantly, the business did not simply conclude that its operators needed to work faster; instead it examined how skilled machining time was being used.

Machine operators were regularly leaving equipment to perform necessary supporting activities. HadFab modelled the impact of introducing dedicated support roles and identified an opportunity to improve utilisation by approximately 20%.

The analysis ultimately highlighted more than 400 hours of potential production capacity.

It also helped HadFab build the business case for additional equipment where the data showed a genuine production constraint.

Read the case study in full >

Maintaining output while removing a night shift

At another one of our customers, who produces doors, production data gave the team much greater visibility of how its CNC equipment was actually being used.

By understanding where time was being lost and focusing improvement activity on the right areas, the manufacturer increased performance on key production lines. The result was significant enough that the business was able to remove its night shift while maintaining production output.

That is effectively a capacity improvement viewed from the opposite direction.

Rather than using better productivity to produce more, the manufacturer used it to produce the same amount with fewer operating hours.

Read the case study in full >

The relationship between capacity and delivery performance

Recoverable capacity matters for more than simply improving utilisation. Manufacturers also need enough production headroom to deal with the variation that is part of day-to-day factory life.

Production rarely runs exactly to plan. Machines break down, jobs take longer than expected, materials arrive late and operators are sometimes unavailable. Priorities can also change at short notice, particularly when an urgent customer order needs to be pulled forward or another job slips behind schedule. Individually, these disruptions may be manageable. The problem comes when a factory is already operating close to the limit of its effective capacity.

When there is very little headroom in the production system, even relatively minor disruption can quickly start to affect delivery performance. A few hours lost on a bottleneck machine, for example, may not only delay the job currently being produced; it can create a queue of work behind it that becomes increasingly difficult to recover.

This is why production capacity has a direct relationship with On-Time-In-Full (OTIF) performance. A manufacturer with some controlled headroom has greater flexibility to absorb disruption, recover lost production time and respond to changing priorities without immediately putting customer delivery dates at risk.

Image: Listers increased machine uptime on existing machines, which enabled it to produce an additional 250 frames per week after deploying FourJaw.  Read the case study. 

That does not mean manufacturers should deliberately accept poor utilisation or keep expensive machinery idle unnecessarily. The aim is to understand where capacity exists, why it exists and whether it is supporting the wider production system.

There is an important difference between unused capacity caused by poor performance and capacity headroom that is deliberately available to protect throughput and delivery performance. A machine sitting idle because it is waiting for material represents a very different situation from a machine with spare capacity, because the upstream process is the genuine production constraint.

Without reliable production data, those two situations can look remarkably similar. Understanding the difference helps manufacturers focus improvement activity in the right places, while also protecting the flexibility they need to deliver reliably when production does not go exactly to plan.

When buying another machine is the right answer

The argument for finding hidden capacity should not be interpreted as an argument against capital investment. Sometimes the evidence will show that the factory genuinely needs another machine, another production line or additional labour.

In fact, one of the most valuable outcomes of properly understanding production capacity can be greater confidence in making that investment.

The key difference, however, is that the decision is based on measured production performance rather than perception. If demand consistently exceeds available capacity, the genuine production constraint has been identified, and that process is already being used effectively, there comes a point where continuous improvement alone will not provide the additional output the business requires.

Before reaching that conclusion, however, manufacturers should understand how much productive time is being lost and whether those losses can realistically be reduced. If avoidable downtime, changeovers, operator constraints or material delays still account for a significant proportion of available production time, adding another machine may simply add more capacity around an unresolved problem.

By contrast, if the constrained process is operating close to its effective capacity, the largest avoidable losses have already been addressed, and customer demand continues to exceed what the process can deliver, investment becomes much easier to justify. The business can also model more accurately what additional equipment is likely to contribute to overall factory throughput rather than looking only at the capacity of the new machine in isolation.

This makes the CapEx conversation much more evidence-based. Instead of saying, “We think this machine is holding us back,” the manufacturer can demonstrate that a particular process is consistently constraining throughput, quantify the production hours being lost or delayed, and show how additional capacity would affect output and customer delivery.

That is a much stronger basis for investment.

The aim, therefore, is not to avoid buying new machinery. It is to make sure that when the business does invest, it is solving the right constraint and adding capacity where it will genuinely increase production.

A simple capacity diagnostic for manufacturers

Before increasing shifts, recruitment or capital expenditure, there are eight questions worth answering.

1. Where is production actually constrained?

Do not assume the busiest machine is the bottleneck. Look at the full production flow.

2. How much production time is genuinely available?

Separate theoretical calendar time from planned production time.

3. How much of that time is productive?

Measure actual machine activity rather than relying on scheduled hours.

4. Where is the remaining time going?

Categorise downtime and delays consistently.

5. Which losses are unavoidable?

Separate necessary maintenance, breaks and process requirements from genuine improvement opportunities.

6. Which losses affect throughput?

Prioritise constraints that prevent orders moving through production rather than simply trying to maximise every machine.

7. How much capacity could realistically be recovered?

Translate improvement opportunities into hours and potential additional output.

8. Does the factory still need additional capacity?

Only then model the impact of additional people, shifts or equipment. This process turns capacity planning from an assumption into an evidence-based decision.

The cheapest production capacity may already be in the factory

Manufacturers will always need to invest.

Growing businesses need people. They need machinery. They need automation and new technology.

But the order in which those decisions are made matters. Adding capacity before understanding how existing capacity is being used risks adding cost without removing the underlying constraint.

Accurate production data changes that conversation. It allows manufacturers to establish how much time is genuinely productive, identify where production is being lost, understand the constraint and calculate how much additional output could realistically be recovered and that's exactly where our production monitoring software comes in. 

Sometimes that analysis will uncover hundreds of additional production hours. Sometimes it will prove that another machine is absolutely necessary. Both are valuable outcomes.

Because the objective is not simply to make machines run for longer. It is to make better decisions about how the factory produces more, protects delivery performance and invests for growth.