Optimizing supply chains and distribution networks is a high-stakes business challenge. The potential rewards are huge and lasting, but the levels of complexity and uncertainty are daunting, and the cost of getting it wrong can be disastrous.

​

Commercial software can help, provided you know what to test, but it doesn’t really help you understand how to represent uncertainty about the future in your decision-making. Planning for scenarios years in the future is a critical factor when you are planning to invest millions of dollars restructuring your supply chain.

 

At Analytics Strategies our approach to network optimization is designed to take account of the real-life challenges, including the inherent uncertainty.

​

First, we take a whole-system perspective. The solution must make sense at the strategic level (where do you build these expensive facilities, so you don’t have to move them in the foreseeable future?), all the way down to the granular operational level (what’s the best way to fulfill this customer’s order?). Small improvements to that strategic level can give you meaningful improvements daily. 

 

Then, we take the mountains of data, figure out the complex interdependencies (often not obvious) and build models that allow you to see the strengths, weaknesses and trade-offs of the various choices you could make. Let’s take two examples of our work.

​

OEM Saves $60 Million

The first case is an industrial OEM that needed to reduce inventory levels and optimize the movement of materials in its supply chains. The company’s three channel partners assembled the final product at their plants located throughout India.

​

The OEM was also concerned that an upcoming change in its industry would mean that the extra overhead of parts movement would invalidate its current supply chains.

​

The input data was extensive, and included cost-to-serve analysis for all customers, value added by the channel partners, supply chain inventories, labor and risk management, evaluating current capabilities and assessing competitors.

​

Analytics Strategies developed a logistics model to evaluate more than 10 different scenarios using operations cost data. In all, we modeled 12 products, thousands of customers, and over five years of demand forecasts.

​

The subsequent business case showed significant potential savings, mostly from transportation, inventory, and facilities costs, after investing to build demand and supply management capabilities.

 

Optimizing their manufacturing and distribution network ultimately saved the OEM over $60 million.

​

Business Boosts Order Fulfillment by 20% to 40%

The second case involved a North American OEM with distribution and manufacturing operations in Japan and China. The business was losing revenue and brand loyalty from its aftermarket service and repair customers because parts and consumables were often not available when needed.

 

In addition, the OEM parts were often much more expensive than their “knockoff” competitors.

 

The company wanted to improve its share of the parts aftermarket without diluting its margins. It also needed to fulfill the demand without too much more inventory, and avoid any conflict with or between partners and their distributors.

​

Analytics Strategies was engaged to develop and iterate a data model to optimize the multiple layers of inventory. 

​

The project’s customer usage data analytics helped the client to anticipate demand for actively managed parts and get ahead of the parts ordering process. Changes to inventory management improved the overall first pass fill from 70% to 92% without any significant increase in inventory investment.

​

Overall, the project improved order fulfillment by 20% to 40% with the same inventory.