Informed Advantage • Part B

How Lytica became a unique analytics company: Part 6B

Lytica’s product is already progressing along the informed advantage productivity journey with pricing answers.  Now we are simplifying and enhancing this product through technology and innovation initiatives while also adding new features and new products.

As previously stated, our Advanced Technology Center (ATC) is using Artificial Intelligence (AI) to enable dramatic change in our technology development and product creation. These new and enhanced capabilities allow customers to deliver profound improvement in management of their costs, security of supply and compliance matters.

Lytica simplified price negotiation with the creation of (FBDC). Through FBDC, you simply upload your data and receive a report – that tells you which components you should renegotiate to a market appropriate price. Our tool benchmarks your pricing to peer market pricing and predicts a suitable price for each component needing attention. You don’t need to waste time negotiating components where you are already priced with a market advantage; you focus only on those with opportunity. What once took weeks of preparation is now days – that’s a huge productivity gain. You eliminate steps and significantly reduce time while increasing the amount of cost savings realized. Yet even with all of these benefits, Lytica can see how enhancements will take productivity to the next level.

FBDC has not eliminated all productivity starving steps from the price negotiation process. The customer’s data preparation remains a productivity constraint.  Many preparation steps that include gathering and analysing data on demand and prices, making assumptions about what’s happening in the marketplace, weaving in expectations from your executive and accounting, target setting for commodities and components, deciding on strategy, figuring out how you might create pricing leverage from Off Approved Manufacturers List (AML) sourcing and a number of other actions which take weeks to complete will now be eliminated. Previously, after doing all of this and completing negotiations, clients remained unsure if they got all of the cost reductions deserved. We have solved preparation and price uncertainty problems, but getting raw data out of a customer’s system to fill in the upload template often remains extremely frustrating for some clients. Simplifying this data preparation step is a focus of our 2018 program for customers who want to work closely with us.

Another area of uncertainty for clients is manufacturer and component risk. Work at the ATC has already revealed hidden information useful in assessing component, manufacturer or supply chain risk. Our continued efforts in the program will result in a 2018 product release.

In the short term our 2017 work creates a step function increase in priced component match rates, untangles data and simplifies the analysis of a huge amount of customer data that has been rejected by our first-generation practices. This data contains MPN spelling errors which have prevented proper match group formation along with pricing anomalies that were hard to decipher.

Our initial AI development objectives were to reduce the production time of reports to 1 hour, increase our match rate to nearly 100% and improve our prediction accuracy. Our prototype developments are already yielding performance akin to these expectations. The 1 hour report should be possible this year and the match rate gap to 100% should be cut in half by then. Where our engineers work 8 hours a day, 5 days per week, AI works 24 hours per day, every day and in about 8 seconds accomplishes what humans need hours to do. This is a two thousand times productivity gain.

I had previously reported on CCE accelerations where analysis on a large file was reduced from 10 minutes to 9 seconds yielding identical results. Some of you have also reported back on the acceleration you see using this tool. Additionally, we have tested prototype cleansing on a customer’s BOM, achieving a 2 minute turnaround on that which took months to complete manually. Finally, we have three projects in progress to improve priced component match rates, one of which, based on MPN cleansing, is showing the potential for a 20% reduction in unmatched components.

Artificial Intelligence applications in broad use today include things like document search, read and compare, spell checking with correction and data analysis such as clustering. These and other items make possible the ordering of random information and access to knowledge that has been hidden within the sheer volume of literature and other content available today. Turning AI loose on our internal component price database and on the world’s content at large has – and will – uncover new insights and knowledge that our customers can use to strengthen their supply base and reduce costs.

As with any research and development activity, success is not assured but our ATC allows us to be disruptive and apply leading edge approaches to our customers’ problems. As stated in previous blogs, the ATC’s purpose is to enhance our customer experience through more timely access to more complete and accurate information on electronic component cost, security of supply and compliance.

Ken Bradley is the Chairman/CTO & founder of Lytica Inc., a provider of supply chain analytics tools and Silecta Inc., a SCM Operations consultancy.

Ken Bradley
Ken Bradley

Ken Bradley is the Chairman/CTO & founder of Lytica Inc., the world’s only provider of electronic component spend analytics and risk intelligence using real customer data.

Articles: 115