UK: +44 (0)1223 264428
USA: +1 (650) 798 5134
by Jez Clements
Most people assume that the products they buy every day are manufactured on production lines that are totally automated – either using super modern, dexterous robots or dedicated machinery.
In reality most manufacturing relies heavily on manual operators. Hence much of the UK’s manufacturing requirement moved to Asia in the 1970s and ‘80s were the labour rates where a fraction of those in the UK. Industries that avoided this trend included;
- Process manufacturing where automation could achieve high utilisation and transportation costs
- Timescales that made overseas manufacture uneconomic,
- Where production required specialised high technology content that’s only available locally.
Today the economic balance of manufacturing is changing again as China becomes increasingly expensive and it potentially becomes cheaper to manufacture in Europe due to several factors; production volumes cannot amortise the set up cost, management overhead becomes too great, or if automation can be justified so labour is minimised. The Asian economies are reacting to this change by investing heavily in skills and technology, so there may be more interesting changes ahead.
While many companies might wish to automate to increase efficiency and improve yield, designing machines to achieve these critical processes is not always easy or straightforward. If all processes were easy to automate, it’s likely it would have happened! Businesses also have to judge that investment in automation will provide a financial return, and not simply end up in expensive scrap.
This is particularly challenging when an innovative new product is being launched. There is a need to plan for high production rates to meet expected demand, but the cost and development time of automation is often high and can be a major constraint if late stage changes occur or the market does not respond as expected. When automation developments fail it can often be put down to not considering the original problem in a broad commercial context, not truly understanding the detail of the assembly operation, or not planning for and dealing with the significant risks early.
A rigorous and robust way of assessing automation opportunities is needed. This can be based on risk based planning where all the steps to transfer to manufacture are considered alongside the product design process. Every project is different but usually you can approach the task from two directions:
• A process that facilitates a complete understanding of a specific automation challenge.
• A higher-level approach that allows a broader problem to be solved or more innovative leaps to be made, that may involve other manufacturing approaches.
To navigate between these paths you need to understand the true drivers of the desire to automate, and the current and future constraints of the system.
When you start to compare humans and machines you really appreciate the incredible flexibility, cognition and dexterity that these skilled human users represent. I’ve spent a lot of time observing manual production lines for medical devices, and the skill level of operators should never be underestimated. Humans are inherently good at making life easy for themselves and often have refined assembly techniques subconsciously. The subtlety of small ‘work-arounds’ that have been adopted or revealed through comments during observation can be incredibly informative as to what the real challenges are, or they can even present potential solutions. For example, an operator assembling a medical device revealed that they looked at the reflections in a part to achieve subtle alignments, this insight enabled us to develop an innovative automation solution to further enhance the process for the operators.
The higher level approach can be referred to as ‘outcome driven’ or ‘jobs to be done’ thinking. This point of view allows consideration of what the ideal final result is without getting tied up with the detail involved in the existing process. For example, a colleague keeps reminding me that if you analysed human gait and tried to replicate it you would never invent the wheel. This approach may not even result in the development of automation, but merely eliminate existing steps for greater efficiency or identification of a completely different design solution. It can help blend design and manufacture constraints, by identifying design aspects which are also key to the ease of manufacture, above the basic product functionality. For example, if you just followed the design philosophy of replicating a manual process step by automation you would be unlikely to end up with a vibratory bowl feeder.
Automation projects are usually driven by the need to improve the reliable categories of time, cost, and quality, but the classic trade-off between them always exists. For example increasing the speed of an operation can allow produce to reach supermarket shelves in a fresher form, whilst improving quality and reliability is critical to implanted medical devices. The needs of the overall project and its market uncertainty often drive the automation roadmap. An optimised program can involve a short term, flexible low-risk ‘get something done quickly’ approach to mitigate market and technical risks while multiple phases of long-term more dedicated systems are developed when uncertainty is reduced.
Automation of a process is often constrained by the changes that are possible to the current system or device. Any change to a regulated medical device could require expensive re-approval. The classic challenge is trying to automate the handling and processing of a device without adding or removing features.
The spectrum between the two basic approaches described here represents options to the development team that manage the risk in the product development process.
At Cambridge Design Partnership our ‘Potential Realised’ innovation process starts with defining the business objective and ends with creating a manufacturing capability that delivers the new product to market at the cost, speed and quality required. To achieve this our team of manufacturing engineers have many years’ experience creating new product supply chains, novel manufacturing processes and automated manufacturing systems. For more information please contact Jez Clements on +44 1223 264428 or firstname.lastname@example.org.
AI can identify factors affecting sports performance that are impossible for us humans to detect.
12 March 2018
The three good practices that help teams move fast on healthcare design projects.
05 March 2018
Stay up to date with all our work and our latest news by signing up to our newsletter.