There are numerous factors that contribute to the success or otherwise of an organisation’s sourcing strategy; choice of suppliers, logistics, legal structures, industry, people, markets served, location, and more. Relationships with these ‘influencers’ are usually well established, having evolved over time as the organisation grew, matured, and sometimes diversified. An organisation that has invested thoughtfully in its sourcing strategies produces better, more profitable products and as a result grows faster than its competitors.
In this article Adam Fletcher, Chairman of the Electronics Components Supply Network (ecsn), suggests that procurement professionals throughout the global electronic components market need to be proactively
involved early in the planning of a new design (or possibly re-design) as its success in the market is almost certain to be dependent on the best available components. But the highest performing devices are very likely to be ‘proprietary’ rather than ‘multi-sourced’ and be in great demand, so the choice of and relationship with supply partners has become increasingly critical to its future success.
Unfortunately, in many organisations, the first time a procurement professional(s) gets involved in the design and development of a new product innovation is when they are sent the bill of materials (BOM) required to produce it!
The engineering department will hopefully have already discussed the aims and system requirements with current and potential suppliers’ field application and sales engineers to glean the latest news and product availability. They will then apply the relevant aspects of their own extensive engineering skills and design experience and discuss the validity of a new design with their peer engineering group and software development and quality assurance teams before making a final choice of component(s).
Even a relatively small OEM (original equipment manufacturer) could be managing 30,000+ components in its in-house inventory at any one time, so there’s always pressure on engineers to design components
they already have approved and in inventory into the new product. A key aim is that the components specified should, where possible, benefit from second-source availability.
Multi sourced components availability is always preferred by procurement professionals because it provides the flexibility to negotiate the most favourable terms, pricing, and theoretically at least, mitigate any supply issues. However, the trend over the past 15 years has been much more towards proprietary components than was historically the case.
Components historical design strategies
Advances in semiconductor technology drives the entire electronic systems marketplace and in turn drive the further development of most other electronic components. Advances in semiconductor technology in recent years has seen a huge increase the integration of previously external discrete components into the IC, significantly increasing its functionality whilst reducing power consumption and cost.
In particular, microcontrollers and microprocessors are often the core of any new OEM product design, but these devices require a whole range of support ICs, passive, interconnect, and electro-mechanical components to perform well.
Invariably designed to meet the identifiable needs of a key customer or new market segment, a new semiconductor innovation often broadens out over its life cycle well beyond the market or volumes
considered realistic at the design stage to become a ‘merchant market’ product, used by numerous customers in a wide variety of applications and is often second-sourced as it matures to
volume production.
A fast-changing market
Each generation of semiconductor technology develops its own bespoke ecosystem to keep up with the rapid ongoing developments and expanding needs of the market. Microprocessors and microcontrollers have transitioned from 4- to 8- to 16- to 32- to 64-bit density into Bit-Slice, RISC, NPU, and now AI devices. Each generation offer design engineers an array of interface and multiple computing cores, peripherals, BUS structures, and memory options. Desirable features such as reduced power consumption and shrinking footprint have also prompted advances in associated memory (SRAM, DRAM, EPROM, Flash, HBM…), power products, interfaces, programmable logic etc.
Time was when the southern shore of San Francisco was the centre of semiconductor innovation often the result of ‘brain-storming’ sessions over a beer in the pub or in meeting-rooms by the Stanford
University community in Palo Alto. Later on, the movement of these innovators between established semiconductor manufacturers or rapidly growing startups led to wide dissemination of their new ideas.
Whilst there was competitiveness among semiconductor manufacturers in Silicon Valley in the ‘60s, ‘70s, and ‘80s, most of the executive management of these companies were scientists/engineers who realised that this ‘collegiate’ approach to solving real-world problems would benefit both their operation and their fledgeling industry as a whole.
Licencing and multiple sourcing
In the late 1980s, semiconductor companies began to adopt more formal business methods. Many manufacturers started to cross licence their semiconductor IP and production technology to competitor
companies.
Producing products for each other increased semiconductor fab utilisation, product availability, and customer confidence and so drove up overall demand for the new product. Cross licencing also
increased manufacturers’ profitability as it enabled them to share back end test and packaging operations.
Sometimes these agreements extended to third or fourth sources, which was something of a headache for the distributors who featured more than one of the manufacturers on their line-card. Under increasing pressure from competitive suppliers to prioritise sales of devices with their logo, authorised distributors sometimes established quasi autonomous subsidiaries to keep the lines apart, but not always entirely satisfactorily!
As technology and the market evolved and matured fierce competition for the available business resulted in significant price and profitability erosion, eventually leading to a significant upswing in M&A (merger
and acquisition) activity between semiconductor manufacturers (and authorised distributors) under pressure from shareholders to achieve the cost reductions that come with increased economies of scale. The
outcome was an exceptionally intense and costly commercial environment, where the return on investment (ROI) experienced significant volatility.
This volatility was amplified, if not primarily caused, by the variations in global economic cycles that directly affected investors’ financial confidence, especially during recessionary phases of these cycles.
EDA and SOC
Parallel with semiconductor hardware developments, the electronic design automation (EDA) industry has evolved and grown substantially offering customers a wide range of options and value points. The cost of entry to EDA is now relatively low but requires significant specialist engineering skills to fully utilise and stay current with this rapidly evolving technology.
Systems-on-Chip (SoC) devices enable customers to design bespoke yet cost effective semiconductor solutions that provide a high degree of product differentiation and importantly, supply control. SOC integration
at technology node level is today available to the masses even at modest production volumes, so it’s no surprise that SOCs are currently one of the fastest growing semiconductor market sectors, encouraged in no small part by third-party services organisations who specialise in the design, development, and testing processes.
Customers however must make considerable up-front investment to achieve the lowest cost of SOC ownership, and accurate forecasting and planning is essential to ensure that their SOC inventory is managed
effectively throughout the life cycle of the end-product. However, like the customer’s printed circuit board (PCB) a SoC is a proprietary solution. The design IP is either owned by or licenced to the customer, not the
manufacturer, which again provides design security and competitive advantage.
Over the past 20 years there has been a very significant reduction in the number of semiconductor companies manufacturing older discrete semiconductor products in a very mature market, but legacy semiconductor technologies can often meet or exceed OEM performance requirements today and can be a financially compelling solution, at least while they remain available from multiple foundries. Unit pricing
is however likely to increase as the volumes manufactured decline.
Support existing suppliers
Serving over 98% of the customers in the electronic components market, manufacturer authorised distributors have a very wide view of ‘what’s happening’ on the ground in real time and are therefore best
placed to provide the ‘intelligence’ about new-generation or legacy components that enables them to be confidently prioritised for new designs. They routinely keep procurement professionals and design
engineers abreast of the best available information about the projected life cycle of a particular electronic component, its current and projected primary market drivers, and the size of the customer base. When production of a component does come to an end, authorised distributors work with the manufacturers they represent to support their customers, advising them well in advance, offering them last-time-buy opportunities and on their behalf holding and managing consigned inventories of end-of-life components.
Concluding thoughts
The direction of travel for electronic components markets is unquestionably towards the greater use of ‘proprietary’ (single source or near single source) components driven by the huge cost of both R&D
and manufacturing technologies required for successful innovation.
When selecting electronic components for their new product design engineers today should routinely consider a device’s availability and lifecycle alongside its electronic performance and package
type. They should also consider future-proofing their new equipment by designing modular PCBs that allow components or sub-assemblies to be easily swapped out if required.
Most procurement professionals have already embraced this new reality and engage with both existing suppliers and interesting, new suppliers each time their organisation embarks on a
new design project. Their experience in negotiating long-term supply agreements is invaluable and their involvement throughout the project will support and guide the process from inception to end of life.
I encourage all partners in the electronic components supply network to actively and positively engage with their partners to negotiate and agree successful outcomes for all. It costs virtually
nothing but the engagement with partners is always beneficial and enhances the performance of all the players in our industry.
This article originally appeared in the July/August issue of Procurement Pro.
Adam Fletcher, Chairman, escn
