To failsafe is to prevent mistakes turning into defects by implementing measures and limits into processes and procedures. The concept was formalised by Toyota in Japan to ‘Poka-yoke’ which literally translated as ‘failsafing’ or ‘mistake-proofing’.


Chelton prides itself on its low field failure rate and its reputation as one of the leading avionics manufacturer for quality and impressive Mean Time between Failure (MTBF) stats. Continuous improvement and striving for perfection is in our blood which is why we design and build failsafe technology. You may just see ‘failsafe’ as a buzzword; empty business jargon or marketing lingo which in reality means nothing. But this is not the case with Chelton, we poka-yoke because failure is not an option.


We caught up with our Head of Manufacturing and Engineering Operations, Neil Kanagaratnam, to take apart what ‘failsafe’ really means for Chelton and how our processes ensure only the highest quality products are sent to our customers.



Robot Cutout From Right
Chelton: Can you explain what failsafing means to our engineers and designers and what you instil in them?

Neil: Our ethos is very much on Simplicity, Repeatability and Continuous Improvement. By applying this to our processes, we are ensuring our customers get only the best systems. We are implementing a number of new processes based on our learnings from legacy products which are simple and repeatable.

Chelton: What are some of the processes you are introducing and how will they achieve their ‘failsafe’ status?

Neil: We are implementing five major processes that will ensure the long-lasting performance of our products.


  1. Zero Defect Plan (ZDP): A well-established process in Quality Management where we ensure, through our process, we ‘do it right the first time’ avoiding waste and costly and time-consuming fixes later on in the process. This way, nothing falls through the net.


  1. Automated Inspection (AOI): To inspect components for faults and categorise them. We have transformed this capability from less of a monitoring tool to more of a process improvement practice by integrating it into our robot arms, Samson and Delilah. These state-of the-art robots have the ability to make decisions on components based on a limits from Sensor Technologies. Anything which is not deemed as a ‘Pass’ based on the criteria the robot sets from data we provide, will fail and will be rejected from proceeding. This way we are minimising the number of wastage later on in the process through early intervention. This will initially be used for larger platforms we provide equipment to but is a procedure we will implement across the whole portfolio in the future.


  1. Digital Twin: Imagine if you could load yourself into a digital version and simulate different scenarios you could go through to achieve a task. This could be anything from understanding how you would cope flying a supersonic jet to the edge of space or how you would cope in different stress situations. You would essentially be creating a digital twin of yourself and that’s what we do at Chelton. From the very beginning, we design and create a digital version of a product and simulate different manufacturing scenarios which bring the product to life but without the wastage of having to build and manufacture a product before knowing whether it would perform as expected. It hugely mitigates error risks and controls the whole processes throughout the final lifecycle so, again, nothing falls through the net.


  1. 3D Model Base work instruction: Don’t you just hate it when you buy a piece of furniture and all of the instructions are in different parts or booklets? It leaves a lot of room for error and you’ll just end up with a wonky and faulty wardrobe at the end of the day! 3D Model Basing is a new technique we are introducing which provides all the steps for build in one, standardised and visual location. The right instructions are all available in one place meaning anyone can build parts easily. This simplifies the entire process, reduces cycle times and reduces wastage.


  1. Six Sigma: Process Improvement is key to failsafe technology and we are driving new developments with the implementation of Six Sigma practices. We do this by embedding people highly skilled in statistical process into the team which allows us to design and develop our products based on unquestionable, tangible statistical information. With this, we are automatically reducing the error rate and making sure the process is repeatable with the same result each time.
Chelton: Your design process seem very robust, is there anything else you do with the scrap and rework waste after it is rejected?

Neil: It might be classified as scrap but to us it’s a treasure trove of mystery to dive into. We don’t just through failed parts away here – we take the opportunity to learn from them and make sure corrective actions are fed back into the very beginning of the design. We use many different precise strategies including analysing microstructural capabilities, AOI, electronic testing, thermal imaging and determining failure modes with a dedicated team recording a database for continuous improvement.

Chelton: Could you elaborate on how the failure rate of a product or part are calculated and why customers should be assured that we are failsafe?

Neil: As with many in our industry, we measure failure through Mean Time between Failure (MTBF) rates which is the average time between system breakdown and maintenance. Our MTBF rates are intimidating to compare against other competitors because our field failure rates are so low and more often than not, it can be attributed to misuse of our system rather than the functional integrity because of all the fail-safe processes we have in place. The only challenge we face is with the internal failure rate because we are screening so much and to such a stringent point to make sure only the best goes to our customer. But the ZDP will lower this moving forward.

Chelton: How will the robots be able to make their own decision and could they potentially miss something a human would have picked up on?

Neil: Our robots are technically cobots so they will work with our human engineers as a team. The robots will act as a brain and the engineers the body to share Chelton’s data driven culture. Engineers will input control and data gathering methods such as failure modes, inspection records, and statistical control and test data, into the robot for it to decide the pass/fail criteria which also means that all of the data is traceable down to the smallest component.

Chelton: What will our customers gain from Chelton having an automated process in place?

Neil: Precision, repeatability and limited human error. A textbook example of this can be showcase in our Laser Tuning capability. At present, everyone in the industry tunes antenna radiating elements manually and, no matter how careful they are, if they take off too much material, the antenna will need to be scrapped. Instead our customers will benefit from our Laser Tuning ability where the robots will decided, based on test data, how much material to take off and instruct the laser tuning process to carry this out. It’s is much more precise than the human hand and eye and will be consistent for mass batches.

Chelton: Thank you for your time Neil, could you sum up in a few words why and how Chelton are ‘failsafe’ and don’t just use it as a meaningless buzzword to describe ourselves?


Neil: Chelton’s processes are Simple, Repeatable, Data-Driven and captained by Continuous Improvement. You won’t find many other avionic manufacturers who can actually live up to and provide evidence for being failsafe.


Neil Kanagaratnam

Neil Kanagaratnam

Head of Manufacturing and Engineering Operations

Back to Latest Thinking