Network Design and Multi-echelon Inventory Optimization

Inventory has been and continues to be the lifeblood of supply chains. Properly managed, it drives revenue and efficiency for companies. But as the nature of supply chains changes, so must the policies used to manage inventory. Traditional inventory management practices are being made obsolete by increasing global supply chains and contract manufacturing, more dynamic product life cycles, and multi-channel distribution.

Figure 1 illustrates the six major components of a technology-enabled inventory management strategy. For each of these components, companies need to identify what levers to adjust within the lead-time, variability and profitability dimensions to simultaneously improve their customer services levels and reduce inventory. Each of these six areas can deliver significant value for companies and help them move toward using inventory management not just to control costs but to drive market advantage.

In this article, we will look at two specific technologies — network design and inventory optimization — and document the usages of these by end users in various industries.

Network Design

Frequency of Network Design Evaluation

Given globalization and dynamic customer requirements, it is concerning that half of companies are looking at their network design only once every two to five years (Figure 2). One reason is the lack of adequate technology tools and associated processes. This area has been dominated by spreadsheet-based tools as well as private consultants.

What Are Companies Doing Within Network Design?

Companies are increasingly using network design beyond just identifying where to build facilities and warehouses. They have started thinking about where and how much inventory to place, along with how to redesign distribution networks to mitigate transportation time, cost and capacity constraints.

Best-in-class companies are more likely than their peers to maximize the use of network design technology:

  • Changing network to support business growth (59 percent versus 37 percent of all companies)
  • Supplier network change modeling (e.g., low-cost country sourcing) (73 percent versus 39 percent of all companies)
  • Changing network to improve transportation costs and capacity (64 percent versus 46 percent of all companies)

    The network design architect at a large diversified manufacturer with global operations indicates that in last 12 months the company has been facing more and more challenges with getting data due to rapid globalization of its supply chains – cost/pound, cost/carton, lead-times, freight lanes, etc. It is resorting to doing simulation and stochastic analysis to come up with more accurate network designs.

Inventory Optimization

Aberdeen research finds that companies have been slowly but steadily moving toward more advanced inventory optimization techniques like multi-echelon inventory optimization. Today, 20 percent of companies report using multi-echelon techniques compared to 12 percent of companies in 2004. And 29 percent of large companies say that they are using this approach. Best-in-class companies are more than twice as likely than other companies to use multi-echelon techniques.

The various approaches used by companies are:

  • General rules-based approach: Set blanket inventory targets for product lines or facilities, frequently as a static “weeks of supply” rule.
  • ABCD approach: Categorize inventory into fast, frequent and sporadic movers and apply inventory targets to each category.
  • MRP/DRP and APS approach: Use a deterministic approach to compute inventory targets, which are computed sequentially for each supply chain echelon (tier); assumes a normal demand distribution.
  • Multi-echelon inventory optimization approach: Use a stochastic (probabilistic) approach to compute inventory targets in one pass across multiple echelons. This approach accounts for supply chain variability (e.g., real demand distribution or lead-time variability) and the interdependencies across multiple echelons.

However, the use of multi-echelon approaches is still maturing. In fact, around 40 percent of the companies that report using multi-echelon tools are, in fact, not taking a pure multi-echelon optimization approach. Instead, they are setting customer service levels for each level in the supply chain and separately calculating the inventories rather than obtaining the inventory based on global cost and service-related calculations.

In addition, 60 percent of companies are planning at the finished goods regional distribution warehouses, and 53 percent are planning at the manufacturing and assembly locations (Figure 3). In other words, very few customers are planning at every level in the echelon. Still, the ability to consider variability and the ability to focus on the critical levels of their supply chain based on their industry (manufacturing intensive or distribution intensive) results in better performance through usage of multi-echelon solutions.

How Old Is Your Inventory Policy?

Another drain on financial performance is that companies’ inventory strategies are rarely kept up to date with real-life conditions. Nearly 50 percent of companies (both distribution-intensive as well as manufacturing intensive) say they update their inventory strategies on an annual or less frequent basis. This frequency of analysis is not sufficient, given today’s global sourcing and contract manufacturing strategies, which create more variability. Survey results show that companies that are above average inventory performers are more than 2.5 times as likely as other companies to update their inventory strategies and policies multiple times a year.

Take part in new Aberdeen research on demand management practices by taking this survey, please click on the following link:

About the Author: Nari Viswanathan is research director in Aberdeen Group's Supply Chain & Logistics Practice. Viswanathan specializes in order-to-delivery and sales and operations planning processes.