When Does Production Automation Become Profitable? How to Calculate the Cost of Delaying Change
TL;DR The profitability of automation is determined by a data-driven economic calculation, not by technology trends. Four decision-making metrics are payback, ROI, NPV/IRR and TCO, each answering a different management question. The cost of delaying change consists of four items: rising machine costs, increasing wages, lost throughput and the cost of capital. Poland has 81 robots per 10,000 employees compared with 231 in the EU-27 (IFR, 2024), indicating an investment gap at a systemic level. MJ Group takes the process from audit through data and modelling to decision-making, followed by implementation and service. Below you will find a ready-made calculation framework and explicit thresholds at which automation should be postponed.
When Does Production Automation Pay Off? The Break-Even Point in Numbers
The break-even point for production automation is the point at which the discounted benefits of implementation exceed the investment outlay and maintenance costs over the solution’s entire life cycle. It is not a single number, but a set of metrics, each answering a different decision-making question.
Payback is the ratio of investment outlay to annual savings. It answers the question of how quickly the capital will return to the company and serves as a quick screening tool. ROI expresses the return relative to the investment as a percentage and makes it possible to rank options of different scale. NPV (net present value) and IRR (internal rate of return) account for the time value of money and form the basis for management decisions. TCO (total cost of ownership) includes purchase, integration, service, energy and training, making it a tool for selecting a supplier rather than evaluating the investment itself.
| Metric | Question it answers | When to use |
|---|---|---|
| Payback | How long until the capital is recovered? | Option screening, initial selection |
| ROI | What is the return relative to the investment? | Ranking projects of different scale |
| NPV / IRR | Does the project create value at the given cost of capital? | Management decision, justification to the management board |
| TCO | What does ownership and operation cost over the life cycle? | Supplier and solution configuration selection |
Typical payback periods for automation projects range from 12–36 months (industry data, 2026). Values below 24 months are usually accepted without additional strategic analysis, while periods above 36 months require non-financial justification, such as staff availability or quality requirements.
The scale of the challenge is measurable. Poland has 81 robots per 10,000 employees, while the EU-27 records 231 and Western Europe 267 (IFR, 2024). The gap versus the EU is approximately 2.8 times. In practice, this means that a plant postponing the decision is competing with an environment that has higher unit productivity.
How to Calculate the Cost of Delaying Change – 4 Components
The cost of waiting is the sum of the benefits the company loses in each year of delaying the decision, increased by the rise in future investment outlays. It consists of four items.
1. Inflation in machine and integration prices. Equipment and system integration service prices rise with demand and component costs. For the model, an estimated 3–6% per year is assumed (estimate, to be verified against market quotations). With CAPEX of PLN 500,000, this means an additional PLN 15,000–30,000 in outlay for each year of delay.
2. Wage growth. The average gross salary in industrial processing is approximately PLN 8,982, while in the enterprise sector it is approximately PLN 9,509 (Statistics Poland, 2026). Each year of delay increases the cost of the work that automation would replace, improving the future business case for implementation while also increasing the current cost burden.
3. Lost throughput and margin. The limitation resulting from the capacity of the existing line means production that was not completed and margin that was not generated. This is the item most often omitted from calculations because it requires estimating unmet demand.
4. Cost of capital. The NBP reference rate is 3.75% (NBP, 2026). Investment models use a WACC of 8–12% (estimate, to be verified), which means deferred savings should be discounted rather than added up in nominal terms.
Formula for the annual cost of waiting:
cost of waiting (annual) = downtime cost/h × downtime hours + labor cost × FTEs + quality losses + maintenance of the old system
Key caveat: the same benefit must not be counted twice. If labor savings result from reduced downtime, they should not be included simultaneously under both labor cost and downtime cost. Each benefit must have one clear and unambiguous allocation.
Downtime Cost – the Most Frequently Omitted Item
The cost of one hour of downtime depends on the sector and scale. Industry data indicate USD 2.3 million per hour in automotive, approximately USD 187,000 in heavy industry, USD 25,000–39,000 in general industry and USD 28,000 in food manufacturing (Siemens/Aberdeen, 2024). These values should be treated as order-of-magnitude benchmarks, not as data for a specific plant.
When measurements are unavailable, downtime cost can be estimated using an upper-bound method: multiply unit margin by production capacity per hour. For a line producing 1,000 units per hour with a margin of PLN 5 per unit, the cost of one hour of downtime is PLN 5,000. This method provides an approximate value, but one sufficient for a screening decision.
MJ Group reduces implementation downtime to 2–5 days through hot-swap, virtual commissioning and control cabinet prefabrication. This is a parameter that directly enters the model as an implementation cost item.
Input Data Checklist for the Audit
The more accurate the input data, the easier it is to assess automation potential and calculate its profitability.
OEE and its components: availability, performance, quality
MTBF and MTTR for critical machines
Line takt time and bottlenecks
Defect rate and the cost of rectification
Cost of one hour of downtime or data needed to estimate it
Energy consumption and load profile
Labor cost at workstations intended for automation
Three Numerical Scenarios: Low, Medium and High CAPEX
The table below presents a model based on explicit assumptions in PLN. All values are indicative and require a quotation following the audit.
| Option | CAPEX | Annual savings | Payback | NPV (r = 8–12%, 5 years) |
|---|---|---|---|---|
| MJ Group – retrofit of an existing machine | 30–60% of the price of a new machine | Dependent on scope | 12–36 months | Positive with savings > 25% of CAPEX/year |
| Robotic workstation | PLN 200,000–600,000 | PLN 100,000–270,000 | 1.7–3.0 years | Positive at r ≤ 12% |
| Cobot at an assembly workstation | PLN 150,000–250,000 | PLN 60,000–120,000 | 1.5–3.0 years | Positive at r ≤ 12% |
| Complete production line | From PLN 300,000, often several million PLN | Dependent on scale | 24–48 months | Requires scenario analysis |
Calculation example: CAPEX PLN 450,000, annual savings PLN 270,000, payback of approximately 1.7 years (PiA-ZAP data, 2025). At a 10% discount rate, the project generates positive NPV as early as the second year of operation.
Sensitivity to changes in CAPEX is significant. With a 20% increase in investment outlay (from PLN 450,000 to PLN 540,000), payback extends to approximately 2.0 years. With a 20% decrease, it shortens to approximately 1.3 years. This means that negotiating the scope and using a retrofit of the existing machine instead of purchasing a new machine have a direct impact on the break-even point.
OEE and Energy in the TCO Model
OEE (Overall Equipment Effectiveness) is the product of three components:
OEE = Availability × Performance × Quality
The world-class level is 85%, while a typical plant achieves approximately 60% (LeanProduction, 2024). Moving from 60% to 85% means an approximately 42% increase in effective production time. For a line operating 6,000 hours per year, this means an additional 1,500 hours of production capacity without investment in new equipment.
Converting OEE into monetary value requires knowing the unit margin. If an effective hour generates PLN 5,000 in margin, a 25-percentage-point increase in OEE translates into PLN 7.5 million of additional annual margin potential. This value requires verification against actual demand.
Energy is the second TCO lever. The average electricity sale price on the competitive market was PLN 518.81/MWh in 2024, PLN 458.24/MWh in 2025 and PLN 486.14/MWh in 2026 (URE). The capacity charge should also be included in the calculation, as it can represent a significant share of total cost in energy-intensive plants. Retrofitting drive and control systems can reduce energy consumption by 20–50% (industry data, 2026).
TCO over a 7-year horizon amounts to 1.8–2.5 times CAPEX (EVS, 2026). This means that for a workstation with an investment outlay of PLN 400,000, the total cost of ownership reaches PLN 720,000 to PLN 1 million. Omitting service, spare parts and training from the initial calculation is the most common error that understates the actual cost.
What to watch for in the model?
- Double-counting benefits: the same saving cannot appear in two positions.
- Omitted service and spare parts: typically 10–20% of CAPEX over the life cycle.
- Omitted training and workforce adaptation time.
- Adding savings in nominal terms without discounting.
- Assuming full utilization of the new production capacity without verifying demand.
When NOT to Automate – Boundary Conditions
Automation is not a universal solution. There are conditions in which postponing the decision is rational.
- Small batches and high product variability. When changeovers dominate production time, a fixed workstation generates losses.
- Lack of process repeatability. Automation requires a defined and stable process.
- Lack of measurement data. Without OEE, MTBF and downtime cost, the model is based on assumptions rather than facts.
- Lack of service resources. Maintaining an automated workstation requires competences that must be provided internally or contractually.
Decision Tree
Answer the following questions. Each answer leads to the next step or indicates what needs to be done before automation.
01 Is the process repeatable and stable?
→ go to question 02
→ postpone and stabilize the process first
→ go to question 03
→ conduct an audit first
→ go to question 04
→ postpone or change the project scope
→ go to the summary
→ plan the service model before implementation
How MJ Group Calculates the Cost of Waiting – From Audit to Decision
MJ Group has operated as an industrial automation and robotics systems integrator since 2006, headquartered in Wrocław with a branch in Katowice. Its areas of expertise include PLC controllers and DCS systems, SCADA/HMI, MES/ERP integration, robotics, production lines, AGV/AMR vehicles, retrofit and IT/OT integration taking into account NIS2 and IEC 62443 requirements.
The workflow is sequential: audit, data, model, implementation, service. The audit provides the input data from the checklist above. Based on this data, a financial model is created with explicit assumptions. The decision is based on metrics, not opinions. After implementation, the outcome is measured by comparing OEE, downtime levels and defect rates before and after.
Available financing instruments include the robotics tax relief, with a 50% deduction of eligible costs, available until 31 December 2026, FENG and KPO programmes, and leasing. Taking these mechanisms into account changes the cash-flow profile and shortens the effective payback period.
MJ Group does not guarantee a specific ROI level. The ranges presented in the article are indicative and require a quotation following an audit, as the actual result depends on the process specifics, production scale and assumptions adopted.
FAQ
How much does an industrial robot cost? The price of the arm itself starts at several tens of thousands of PLN, while a complete robotic workstation costs PLN 200,000–600,000 and cobots PLN 150,000–250,000 (industry data, 2026). The final value depends on the scope of integration, grippers, vision systems and safety.
When does production automation pay off? When payback under realistic assumptions is within 12–36 months and NPV at a discount rate of 8–12% is positive. Downtime cost, labor cost and production scale are key factors.
What is OEE? OEE is the product of availability, performance and quality. The world-class level is 85%, while a typical plant achieves approximately 60% (LeanProduction, 2024). An increase in OEE directly translates into effective production time.
How much does line downtime cost? Industry data indicate USD 2.3 million per hour in automotive, approximately USD 187,000 in heavy industry and USD 28,000 in food manufacturing (Siemens/Aberdeen, 2024). For a specific plant, the value should be estimated based on unit margin and line capacity.
What is the typical automation payback period? Automation projects are dominated by 12–36-month payback periods (industry data, 2026). Retrofit usually pays back faster than building a new line because the investment outlay is 30–60% of the price of a new machine.








