In an electronics system design, the final product is only as good as the quality, price, and availability of the components that comprise it. Engineers and sourcing professionals continually work to improve their part selection techniques, since an error in the component selection can cause costly redesigns, or simply kill a product before it even gets to market.
In recent years, parametric search engines have facilitated this task. Narrowing down options by applying tangible filters (figure A) has become an alternative--and a much more scientific--method to vendor meetings and individual knowledge. Search engines do a great job at indexing data fields and returning component information according to specific filters such as the value or tolerance of a resistor.
However, selecting a component after researching and comparing specifications, constructions, and functionalities doesn’t necessarily make it the right part for a product. It certainly makes it the right part for a design. And there’s a tremendous difference between the conceptual stage and the physical item.
When passed to the sourcing or purchasing team, the list of selected components needs to be sourceable and, in most cases, for the cheapest possible price.
Here are a few questions that need to be answered:
- Is each part available in the requested quantity?
- Which suppliers have it in stock?
- Which supplier has the best price?
- Is there an alternative component that matches the desired form, fit and function that is easier to get or cheaper to purchase?
- What are the risk levels associated with each component (end of life, compliance, etc.)?
Finding the answers to these questions requires more than parametric information. Machine learning and artificial intelligence allow a new generation of sourcing tools to identify bottlenecks before they happen and enable proactive recommendations. This is when analytic search delivers incredible value in the product development cycle.
Let’s look at a concrete example.
Looking for a Texas Instruments buffer/driver (SN74LVC1G07DCKT) triggers two warnings:
- No single distributor has enough inventory to match my requested quantity. Digi-Key is the closest but still short of by1,500
- The current price is above market average.