Predictive analytics as a tool for Inventory Management

In between the myriad of applications existing in the market to support part of the supply chain or the entire thing, as an entity on itself or as a part of an Enterprise Resource Planning system, general purpose Supply Chain Management software or Industry/function specific, find a particular topic of interest (for this blog and for the class)  can be a little tough.

Luckily, I found this article from IBM to share: "Predictive Inventory Management: Keeping your Supply Chain in balance"

The ultimate way to accomplish adequate inventory levels, balancing out of stock risk and inventory cost, is almost solved with the proposed predictive approach.

"Successful management of inventory levels is one of the key factors in the financial success of every manufacturer, wholesaler or retailer. As the primary component of working capital for most of these businesses, inventory affects the flow of money through the business as much as it does the flow of materials and merchandise through the company’s operations."

Despite the particular interest in the article to promote the IBM product, the most important thing is the general background provided to support the benefits of using analytics in the inventory management process.  The predictive approach prevent the losses inherent to the widely used reactive approach.  The change in demand (mainly) or other variables through the value chain, triggers adjustments in the inventory level at several stages, but the changes are performed in the aftermath.

The contrasting conclusion provided in the article says:
"Predictive inventory management helps ensure that the business can meet the ever changing needs of a diverse customer base while minimizing costs and end-of-life discounting, and maximizing profitability. An optimum level of inventory, which enables the business to free up working capital, can give a business the flexibility to take advantage of capital-intensive opportunities when they present themselves."

The methodological proposal for a "Predictive Inventory Maturity Model" is a sequence of progressive steps, well depicted in the figure reproduced below:

Instead of reproducing the IBM definitions, I propose 'think about it'.  Do you agree with this model? Is complete and compelling? I'm biased, I got seduced at first glance so, further discussion can sound like advertising....