The purpose of an organization is to enable ordinary human beings to do extraordinary things – Peter Drucker

What Is a CNC Machine Breakdown Investigation?

A CNC machine breakdown investigation is the structured process of gathering accurate data from the shop floor immediately after a machine stoppage — before memory fades and conditions change. Getting the right information at the point of failure is what separates a fast, targeted repair from hours of guesswork.

According to BS EN 13306:2017, the current BSI standard for maintenance terminology, a breakdown is defined as the inability of an item to perform a required function — a distinction that also separates corrective maintenance (restoring function after failure) from preventive maintenance (avoiding failure through scheduled intervention). Understanding which category, you are dealing with at the point of failure shapes everything about how you investigate it.

When a machine stops, the first people on the scene are almost always the operators — not the maintenance engineer. A well-designed set of questions gives operators a framework to make accurate first assessments, decentralise the initial diagnosis, and capture data that would otherwise be lost by the time a specialist arrives.

I saw this first-hand working at AI, a precision defence manufacturer. We had two Goodway CNC lathes running on the production line, both fitted with the same type of chuck. Standard practice was to pull each chuck off every three to four months for a strip-down, inspection, and regrease before refitting.

One day the line stopped. By the time I got involved, the machine had already been down for five hours with no production coming off it. I went to the floor, spoke with the operators and the line manager, and worked through my question framework. It did not take long. The maintenance history questions surfaced the problem almost immediately — one of the two chucks had missed its service interval, seized, and had to be taken off for full refurbishment. No 5-Why analysis required. The cause was straightforward: a missed scheduled service.

The solution was equally clear — purchase a third chuck so that one could always be in service while another was being stripped and inspected on rotation. The only hurdle was the price: £2k. The justification practically wrote itself. While that machine was down, the entire production line was standing. The purchase order was approved the same day. The new chuck took a few weeks to arrive, but from that point forward the rotation system meant the situation could never repeat itself.

That is what a good question set does. It gets you to the answer before you start reaching for more complex diagnostic tools.

Why a Structured Question Framework Matters

Most CNC machine breakdowns can be resolved quickly once the right question is asked. The problem is that without a framework, the wrong questions get asked first — or none at all. Operators default to resetting the machine and continuing, which clears the fault but destroys the diagnostic data.

ISO 13374-1:2003, which establishes guidelines for machine condition monitoring and diagnostic data, recommends a structured data processing and information flow — beginning at data acquisition and moving through to health assessment and corrective action decisions — whether that flow is implemented manually or automatically. The question framework below is the manual, shop-floor-practical equivalent of that principle: a guided sequence that captures the right data at the right moment.

The list is not intended to be used in full for every incident. Select the questions most relevant to the uncertainty at hand and the specific machine or process involved.

CNC Machine Breakdown Questions — Organised by Category

Use this framework at the point of failure. Categories are not sequential — start with whichever is most relevant to the apparent failure mode.

Setup and Pre-Conditions

• What was the condition of the workpiece before loading?
• What is the condition of the floor and working area around the machine?
• Were there any changes to speeds and feeds prior to the stoppage?
• Were any changes made to the machine on the day of the issue before the fault? (Maintenance, environment adjustments, operational alterations)
• Has there been any change to the operation? (New fluids, filters, cleaning, tool changes)
• Was the correct programme, tooling, and workpiece loaded?
• Were the Work Instructions / SOP followed?

Technical Indicators

• Was the machine vibrating during operation?
• Are there any fault codes displayed on the control screen?
• Is the issue mechanical, electrical, or software-related?
• Were there any unusual noises from the machine?
• Has any part of the machine been overheating?
• Are there any unusual smells?
• Do multiple failures occur simultaneously?
• Can the problem be reproduced without loading a workpiece?

Work Sequence and Discovery

• How did the operator discover the issue?
• What was the sequence of events that led to the fault?
• Does the fault occur at the same stage of the process each time?

Maintenance History

• When was the machine last checked or serviced?
• Has any part of the machine been replaced recently?
• Were the scheduled machine checks completed today?
• Are there any notes in the service engineer’s report indicating prior concerns?

Operational Context

• What was the machine doing immediately before it stopped?
• How long after starting the operation did the machine stop?
• Was there any early indication that the machine was not performing normally?

Environmental Conditions

• What is the temperature inside the factory at the time of the stoppage?
• What are the humidity levels at the time of the fault?

Failure History

• Has this specific issue occurred before?
• Has the machine’s performance changed noticeably in the last few days?

How to Convert Machine Breakdown Data Into Useful Information

Capturing answers is only the first step. The real value comes from converting repeated captures into pattern data.

The most effective approach I have used is to design the capture document as a series of tick boxes rather than written answers. This removes ambiguity, speeds up completion at the point of failure, and — critically — produces categorical data that can be analysed statistically over time.

Once you have sufficient records, start with a run chart. This shows you how often the machine stops, at what frequency, and whether the rate is changing. From there, apply the Pareto method to identify which failure categories account for the majority of stoppages.

The published evidence supports this approach. A study published in the African Journal of Science, Technology, Innovation and Development applied Pareto analysis to machine breakdown records from a manufacturing firm and found that 6 out of 11 subsystems accounted for approximately 80% of total downtime — allowing a more targeted preventive maintenance plan to be developed from the data. This is precisely the pattern you are looking for when analysing your own breakdown records: a small number of causes driving the majority of stoppages.

The Reliability-Centred Maintenance (RCM) process, as described in the IEEE literature, is designed to focus engineering attention at the component level in a formal and disciplined manner, using systematic analysis of failure mode, rate, and criticality data to determine the most effective maintenance requirements — with the goal of reducing scheduled maintenance burden while sustaining operational readiness. The tick-box capture system described above feeds directly into this type of analysis: your question framework collects the raw failure mode data; Pareto analysis identifies the vital few; RCM logic then determines the appropriate maintenance response.

A good example of this came from work I did alongside the Quality Manager at Accuracy International, who had developed an NCR reporting system using a Microsoft Access database. Operators could log faults directly at the point of occurrence on a shop floor PC. The system automatically generated a monthly Pareto chart for each fault category, alongside run charts showing when problems were occurring and at what frequency. It gave us something that questioning alone could never provide — a view of patterns across time.

One that stands out involved a stick-on part that was die-cut on a press. Rework was being recorded against this part periodically, but the cause was not immediately apparent from any single incident. There were two die sets in use, and neither looked obviously problematic on inspection. It was only when we drilled down into the data that the pattern became clear — the rework spikes were occurring at regular intervals of a few months, and when we mapped those intervals against the production records, they aligned precisely with the switchover between the two die sets. One of them needed re-sharpening. Without the run chart showing the frequency and timing, and the Pareto identifying the part as a recurring contributor, that pattern would have remained invisible.

That is the value of building a data capture system around your question framework. The questions get you the answer on the day. The data gets you the answer across the months.

Frequently Asked Questions — CNC Machine Breakdown Investigation

What is the first thing to check when a CNC machine breaks down? Before anything else, secure the machine and record the conditions at the point of failure — fault codes, sounds, smells, and the last operation performed. Resetting without recording destroys diagnostic data that cannot be recovered later.

How do I know whether a CNC breakdown is mechanical, electrical, or software-related? Start with the fault code on the control screen, if present. If no code is displayed, work through physical indicators: vibration, noise, and heat point towards mechanical issues; unexpected axis behaviour or control errors suggest electrical or software causes. The question framework in this post is designed to help narrow this down systematically.

Should operators or maintenance engineers complete the breakdown record? Operators should complete the initial record at the point of failure — they are first on scene and hold the most accurate account of conditions immediately before the stoppage. The maintenance engineer then reviews and supplements the record during diagnosis.

How many breakdown records do I need before Pareto analysis is useful? There is no fixed minimum, but a meaningful pattern typically requires data from at least 20–30 separate incidents on the same machine or machine type. For low-frequency failures, this may take several months to accumulate.

Can this question framework be used for machines other than CNC machines? Yes. The categories — setup, technical, maintenance, environment, and history — are applicable to most production machinery. You may need to add machine-specific questions. For a press, for example, you would include questions about die condition and tonnage settings.

Summary

Most CNC machine breakdowns can be identified and resolved quickly once the right data is captured at the right moment. The question framework above is a guided starting point — not a rigid checklist — and should be refined over time to reflect the specific characteristics of your machines and processes.

References

Nowlan, F.S. and Heap, H.F. (referenced in: Smith, A.M.) ‘Reliability-Centered Maintenance’, IEEE (overview paper). Available at: ieeexplore.ieee.org/document/5222285

BS EN 13306:2017 Maintenance. Maintenance terminology. BSI. Available at: knowledge.bsigroup.com

ISO 13374-1:2003 Condition monitoring and diagnostics of machines — Data processing, communication and presentation — Part 1: General guidelines. ISO. Available at: iso.org/standard/21832.html

Abubakar, U. et al. (2016) ‘The Pareto principle and a hazard model as tools for appropriate scheduled maintenance in a manufacturing firm’, African Journal of Science, Technology, Innovation and Development, 8(2). DOI: 10.1080/20421338.2016.1147203

Author Bio

Stuart Bateman is a Chartered Engineer with 22 years of experience across project, manufacturing, and production engineering. He writes about engineering tools, career development, and practical manufacturing techniques for engineers at every stage of their career.

External Lean – Resources

Want to learn more about Lean, then check out these 5 Lean organisation websites that I highly recommend:

Lean Smarts – LeanSmarts.com offers an online training academy designed to support lean manufacturing transformations through engaging video-based courses and certification programs Leanuk. The platform provides ready-to-use training materials with facilitator guides, covering topics like 5S, waste elimination, one-piece flow, pull systems, and visual management based on the Toyota Production System Leanuk. Organizations can deliver training through personal certification, facilitator-led sessions, their own training platform with up to 250 users, or integrate content into existing Learning Management Systems Leanuk.

Lean Academy – LeanUK.org (Lean Enterprise Academy) is a not-for-profit organization that helps individuals and organizations become self-reliant in lean thinking and practice through research, teaching, and coaching Leanuk. They conduct action research to develop learning materials and processes, deliver capability development through their online Lean Learning Platform and workshops, and share findings through events like the UK Lean Summit Leanuk. Founded by lean pioneer Dan Jones, they’ve been involved in developing and spreading lean thinking in the UK for over 40 years across industries including automotive, healthcare, government, and services Leanuk.

CI Tool kit – CIToolkit.com provides professionals, teams, and students with practical tools, methods, and resources for implementing continuous improvement and achieving operational excellence citoolkit. The site features libraries of downloadable infographics, team exercises and games for Lean and Six Sigma workshops, visual illustrations of concepts, and step-by-step how-to guides. It also offers digital products including training materials, templates, and worksheets designed to support continuous improvement journeys citoolkit.

iSixSigma – iSixSigma is a comprehensive online resource dedicated to Lean Six Sigma methodologies and continuous improvement practices isixsigma. The platform offers articles, case studies, and certification guides covering topics like Six Sigma, Lean manufacturing, Kaizen, and Business Process Management. It serves practitioners from beginners to experts seeking to reduce defects, improve efficiency, and drive business results through data-driven process improvement isixsigma.

Lean Production – LeanProduction.com is a free educational resource focused on lean manufacturing techniques and eliminating waste from production processes leanproduction. The site provides easy-to-understand guides on topics like 5S, TPM, OEE, and SMED, helping manufacturers improve efficiency, reduce costs, and increase productivity leanproduction. Brought to you by Vorne, it offers practical tools, templates, and resources designed to be quickly implemented by manufacturing teams looking to achieve measurable improvements in their operations.

---

What are your thoughts? Have I covered everything or is there more you know and would like to share?

I’m always learning and improving this site and my blogs, so please feel free to get in touch with me via LinkedIn or this site to discuss any topics I have covered.

If you’re having trouble finding ways to progress check out these sites filled with free learning tools: