Value Engineering for electronic product development
We sat down with Steven Richardson, currently on the shortlist for Design Engineer of the Year in the British Engineering Excellence Awards, to discuss value engineering techniques that work during the design and manufacturing stages of the electronic product development process.
What is Value Engineering?
Value engineering is a method of assessing your current electronic product design and development, and then using a series of cost reduction techniques to maximise profit or pass on savings in unit costs to your customers. In an age of both chip shortages and a tech boom, using effective cost reduction strategies is one of the best ways to save costs and add sustainability to your business plans for volume production product lines.
First thing’s first – here’s one thing you should do before considering value engineering practices.
Getting A Great Product Out To Market Fast
Gain first mover advantages first
Value engineering is a great solution if you have a working product. If you are launching your first product, then cost reduction should never be your priority. You need to get your product out to the market and add contingency. From this point, you can then see which features are being used and talked about by your customers. This way you can conduct research from user experience and, in some cases, data collection. Consider if you need the bells and whistles to see how people react to it and which features are important.
For example, do you need all those Ethernet ports, flash, RAM and the biggest microprocessor? In addition, you may decide not to implement over the air (OTA) bootloading or firmware updates; you could save a lot of money here.
These all-encompassing bells and whistles products offer the greatest opportunities for value engineering. However, if you are creating a premium version of your MVP you will want to avoid feature creep and other factors as you go along to prevent adding avoidable costs to your design.
Design-Based Value Engineering
Where to Start with Value Engineering
Assuming you have a working product first you can look at where you can make changes. Are users using every feature? If not, you can look at pairing it back, this is the quickest product cost reduction technique.
Start with a Design Review
The best way to work out how to save on the design itself is with a design review.
Assess what’s on the board and its purpose. Then take the BoM (Bill of Materials) to find cost saving opportunities. This is something we can help with.
Design Reviews from Ignys
Start Expensive and Work Backwards
Look at the BoM line by line, order these by the most expensive part first and start there. It’s important to know where your costs are for a good birds-eye view.
For example, there is no point taking 3 resistors off when you can take a whole IC (Integrated Circuit) off as a quick way to save costs without compromising functionality.
Microprocessor Techniques
It often makes commercial sense to use the biggest microprocessor when you are going to market as it avoids crippling your software. However, post product launch, you may find that your customers are only using half or even 1/10th of the chip memory. Often families of processors have lower memory options, so this is a quick and easy way of scaling back.
Software Code Optimisation
Your software code could be optimised. If you have written modular and portable HI2code, there are massive savings that can be made from changing to alternative manufacturers of microcontrollers.
CEM Relationships
Start by having an honest, transparent chat with your CEM. From here, you can find out where costs are adding up. The aim is not to bankrupt your CEM! Don’t cause a hostile exchange with your supplier, instead make it a positive win-win situation for both of you.
Value engineering will ultimately allow you to make units cheaper whilst also making it easier for your CEM to produce a higher volume for you, reduce waste, increase yield and speed up their own processes. This puts you both in the best place possible and could strengthen your relationship. Your CEM will have relationships with major players and component suppliers which gives them more options to source parts than you, often with supported pricing.
Circuit Board Design and Value Engineering
An often-overlooked part of every design is the bare board itself, the PCB. Great savings can be made at bare board level.
Start by looking at the following areas:
- Plating options
- Number of layers
- V-scoring (cheaper) vs milling (more expensive)
- Board size, and utilisation of the FR4 – if you’re aware of what needs to go on the manufacturing and assembly panels, you can get your better utilisation out of the industry standard panel sizes & save money.
Rationalise your board
Start with the bare board you’ve got. Once you’ve found out what your PCB is costing you, you can check the design; do you need gold plating with those fancy features, for example? Does it need that many drill holes?
The PCB can often be rationalised by reducing the number of drill sizes which speeds up the process, 2 passes for 2 sizes takes twice as long. Larger drill and milling apertures can be stacked, allowing multiple boards to be drilled together. So even down to board level you can make huge savings based on these properties. It is all about time and process.
Minimise the time needed for both machines and for people to handle the board, especially the latter. In manufacturing time is money.
At board level you can also consider plating options and number of layers as well as any special features like microvias and back drilling. Can you use V-Scoring instead of milling?
V-scoring is the process of cutting a “V”-shaped groove on a circuit board at the top and bottom of a PCB, like a pizza cutter. Of course, with V-scoring you need to consider the advantages and downsides. If you have components at the edge of the board breaking out the V-scored panels can damage them but it is also a lot cheaper than milling and a lot faster. V-scoring can only be done in straight lines but it’s quick, so if you don’t need rounded edges this can be an effective way to save on cost.
PCB Panelisation
How you panelise your board size can have a big RF5impact because it’s all about utilisation of the Fr-4. When you design your board size, consider the panel size and required items for that panel. For example, using 40% of the panel where the rest is wasted and thrown out is not good utilisation.
If you can get your design to fit / tessellate multiple times in a given panel size there is room for improvement up to 78% utilisation (it is harder to go much higher than this as the manufacturer requires features to be present to work with their processes and machines). This optimisation could give you much higher yield. Consider how much space you need between PCBs. What’s your PCB size, what are the assembly and manufacturing panel sizes required? These are questions I ask whenever I get started on any new design, let alone a cost reduction strategy exercise. Plus, if your board doesn’t tessellate well or is a funny shape, with fancy corners for no good reason or unusual cut-outs, that is going to cost you more money for no added benefit.
Manufacturability and Value Engineering
Assembly consolidation
Regarding assembly consolidation, less is more when it comes to the number of parts.
Do you need a 10/11/12k resistor? If you need 12k for part of it, have a look at the design, it could make sense to make them all 12K. For example, you could have a 10K pull up resistor and a 12k for a specific switch mode power supply.
If you can use all 12K, instead of a handful of 12K and 10K resistors, this can save money in procurement and the sourcing of parts. It also saves time for loading the pick-and-place machine, and with the inspection and programming time. It may be minor, but labour time saved en-masse can make a difference, especially if there are multiple reels involved.
These machines have fast multi-pick heads that rotate and can rapidly collect multiple parts from the same reel and carry them to the board much faster than picking from different reels.
BoM (Bill of Materials)
Consider if every item on your BoM needs to be there. During development it’s easy to add parts you think you’ll need and then forget about them later. Do you need those high precision parts? Are you getting the resolution you need or is there too much noise in your system? Do you really need 24-bit or would 16-bit be enough? Look at the BoM, line-by-line, ordered by expense and target the heavy hitters. The general rule of thumb when it comes to the bill of materials is to ask the following question: are you getting the benefits you need, or are they luxuries you are not making the most of?
Reduced dependency
Reduced dependency on multiple parts is also ideal during the semiconductor chip shortage period. Having less unique line items means less procurement and limits your sourcing pain points. Also, if you order in higher quantities its likely you will have more power talking with the suppliers as bigger players often gain the upper hand through economies of scale and price points.
Assembly time and testability
How difficult is it to test and assemble your product?
Beyond the composition of the printed circuit board, human labour is one of the biggest costs.
If people have to touch it, sit there for a minute and then press a button, that costs money. And for the rest of the time in-between you are paying someone to sit there.
This brings us on to your test fixture.
Optimise your testing solution
Can your test jig be made to test more than one board at a time? If your test fixture solution allows you to test twice the number of boards, you can half testing time. And, if you can test an entire assembly panel with six versions of your board, you can slash that time down even further.
There is of course some added risk when testing an assembly panel before breaking out each individual board, as the board is handled and possibly stressed during this process after testing. A multi-way test fixture will also be bigger, having a larger upfront cost, and the time to test might be slightly longer than that of a single board. However, the more the process is paralleled up, the faster it will go. Ultimately, using less of an operator’s time whilst testing can save a lot of money.
Box Build
How can you minimise your assembly process?
Many CEMs will box build for you. Consider how your PCB is physically attached to the enclosure and what the human factor is in the process. Do you simply place the board in and snap it down with straight-forward plastic clips or do you use 16 manually inserted screws?
Consider ribbon cable vs board-to-board. Where are the I/O (Input/Output) placed? Think about the assembly order and if a connector can be reached when the board is inserted or does that need connecting beforehand? Is the cable long enough?
Is there an awkward assembly procedure with several steps which takes time? If it is a fiddly process they are going to charge you for it.
Job variation
Job variation is also important. If an operator does the same thing day in day out, they will get tired. If they get to work on different sub-assemblies that could influence the time taken. You may want someone building front panels one hour and threading cables the next, this keeps them more interested and by doing one task at a time they will be a lot faster than doing very linear 10-step process all day every day.
Instructions and timings
Clearer instructions equal less errors.
Do you have a standard operating procedure for various aspects of assembly? Consider your assembly and manufacturing instructions… can you provide your CEM with a clearer and quicker way of doing things? You can’t assume that they will magically know the best way to put your product together.
They may also be charging you for something which they claim takes 15 minutes. If this doesn’t seem right you may want to review it with them, but be sure to consider fatigue and different skillsets in the equation. Building a transparent and collaborative relationship with your supplier is key.
Offshore manufacturing
You can also weigh up the pros and cons of using an overseas assembly line. They often come in cheaper even with shipping, however the quality, environmental factors and the fair working policies are also worth bearing in mind.
UK manufacturers have strict rules around fair pay and working conditions and tend to be easier to control or contact in terms of quality. Confidence and trust will play a key factor in whether you choose this option or not. Your volume will also dictate whether going overseas makes more commercial sense.
Should my design and manufacturing partner be the same company?
Storage
Consider how your electronic product design lends itself to transportation – does the product tessellate well? Can it be stored well or does it fall over easily? Form factors are important as well as packaging (bear in mind packaging will affect the environment which some of your customer base are likely to care about). Overindulgence on packaging will hit your back pocket as well as leaving a bad taste in the mouths of your customers.
Know your break point
Every design can be reduced in cost, but is the time spent worth it? Know when to stop. 20-30% off unit cost is often achievable. After a certain number of cost reduction iterations there will come a time when your design and development budget is better spent elsewhere. But before you get to that 30% optimisation there are ways to win in the battle of cost vs design.
Ignys Poppy (Sales left) and Stephen Keetley (second from Right) accepting the design team elektra award
Value Engineering At Ignys
We specialise in helping developers to bring great ideas to life. We have a great number of value engineering techniques that we put into practice for customers. If you’re interested in saving costs speak to our team through a frank discovery call.
Start with a design review
The best way to enter the world of value engineering is to take your current design and get a design review. We can do complete design reviews or partial ones such as pcb design reviews.
Does Ignys get involved in electronic product design?
It depends how much design freedom we get.
We can work with designers through our own methods and trusted partners to help you design a product in a certain way for cost reduction and design ease. Often it happens the other way round, product ideas often stem from the look and feel of the new product, people don’t usually start off thinking about how they will fit a traditional or flexible PCB inside them.
In a world where visuals didn’t matter you could make things twice as big, square and bulky and charge a lot less for design, but the tech industry is moving in the opposite directions for many reasons. Ease of use, portability and the visual aspects of product design all play a big role in the appeal and marketability of a product and often designing around these beautiful form factors is worth it, just ensure that is the case for your product.