Timken solution improves inventory optimization, supply chain performance
Inventory and demand forecasting remain critical challenges in the automotive aftermarket, but technology is advancing that can help suppliers and distributors make better decisions about the placement of inventory – right down to the zip code – that can increase turns, reduce obsolescence and save costs.
Timken is one of several aftermarket companies that is leveraging sales history, registration data, and other information, along with complex analytics, to improve sales and reduce costs. The company uses an internally developed solution called Demand Insight to help distributors optimize their inventory.
Timken's catalog team matches parts and vehicles, and combines that information with vehicle registration and replacement/failure rates, along with internal sales data. Crunching that data using proprietary algorithms helps them predict how many parts will be needed in a given geography, and how those parts sales will fall within the premium aftermarket, economy aftermarket and OEMs.
More than 300 distributors and Timken sales representatives now use the online system, and it includes more than 4,000 parts with vehicle applications. Customers can access detailed analysis and charts showing demand for specific parts based on location. They can also view the most popular parts in order of demand, along with peak demand based on part lifecycles.
"We've been able to take VIO and cataloging data, and add a third element in the replacement rates," says Barry Harris, global manager of automotive aftermarket at Timken. "We're an OE manufacturer, and we know the replacement timing of our parts. Using that information and some sophisticated logic we developed, we can say in general when you can expect demand for those replacement parts to begin. The question then is, how do you make that forecast better? That started us down the path of using geographic information, and now we're even adding information on weather patterns and local road conditions."
Timken launched Demand Insight in 2010 and the Canadian market went online in 2012. There have been other tweaks over time, including the addition of a data source that helps predict new vehicle registrations for up to two years into the future; flags that indicate part applicability for agricultural or heavy-duty vehicles; and the ability to do multiple searches in multiple tabs. In 2014, the company added medium-duty vehicle aftermarket demand.
This type of data-driven, intelligent forecasting is expanding in the aftermarket. Delphi, for example, has an Intelligent SKU Management solution that uses proprietary research, market trends and customer provided information to help optimize inventory for local markets. Goodyear Engineered Products has the DataDrive Market Intelligence System, which uses Veyance Technology's algorithms to evaluate stocking levels by region, county and metro area for warehouse distributors. TRW Aftermarket uses a big data solution from PROS Holdings to make tailored offerings and improve pricing strategies based on inventory data.
Large distributor groups also have developed their own inventory management tools for forecasting.
Increased sales, lower inventory
In 2012, when Timken won the Polk Inventory Efficiency Award, the Demand Insight tool, in combination with new ERP, warehouse management, and logistics improvements, had a tremendous affect on the company's performance. The combined solutions helped boost year-over-year sales by 19 percent, reduced inventory 12 percent, and increased fill rates by 13 percentage points, according to information released by Polk at the time. Even though Timken has to purchase access to some of the information (like vehicle registration data), the benefits provide a significant return for the company and its customers.
Prior to demand insight, the company relied on historical data. When the initiative first began, there were technical limitations to overcome. "The first real challenge we met was trying to mesh registration information with our catalog data," says Tom Muck, principal systems analyst at Timken. "There are a lot of specific attributes, and we had to come up with a solution that took all of those situations into account. Then we applied basic replacement rates to project demand."
"Our crystal ball was pretty dark before, and it always looked backwards," adds Harris. "We've taken some elements that anybody can get, and matched those with the replacement rate, and we saw a lot of improvement in the forecast."
Timken built the solution internally. Both the sales team and customers were initially skeptical of the forecast, but employee training and the system's performance have proven its reliability over time.
"The key here is when you match up the data and tell the customer that they are missing out on 4 percent of the market share, for example, and here are all the part numbers you aren't even stocking for cars that are right in your own back yard," Harris says. "They might be lower on certain products than the data would indicate. They might be pricing it wrong, or not promoting it."
As the system evolved, the data became more granular. "We can provide information down to a certain geographic level," Muck says. "We improved some of the replacement rates to refine the data, and we've gone through several iterations since then."
To keep the system updated, each quarter the company adds the latest catalog information and latest vehicle registration data. "We also have an annual review of the replacement rates that we use for the system," Muck says. "We go back through all the various facets of the system annually, and we have score-carded metrics that we use to make sure the data accuracy is where it needs to be."
"At the end of the day, category management should allow you to reduce inventory internally and with your customers," Harris says. "This is a tool that works well for our customers and for us."
Subscribe to Aftermarket Business World and receive articles like this every month….absolutely free. Click here.