Machine Learning Times
Machine Learning Times
EXCLUSIVE HIGHLIGHTS
Effective Machine Learning Needs Leadership — Not AI Hype
 Originally published in BigThink, Feb 12, 2024.  Excerpted from The...
Today’s AI Won’t Radically Transform Society, But It’s Already Reshaping Business
 Originally published in Fast Company, Jan 5, 2024. Eric...
Calculating Customer Potential with Share of Wallet
 No question about it: We, as consumers have our...
A University Curriculum Supplement to Teach a Business Framework for ML Deployment
    In 2023, as a visiting analytics professor...
SHARE THIS:

9 years ago
Predictive Analytics: Turning Data Into Inventory Insights

 

For more case studies on Predictive Analytics and Manufacturing, check out PAW Manufacturing, June 8-11. 2015 in Chicago.

Picture this: a shop floor manager sits, hunched over the desk. It’s late, long past the shift change but he’s not going anywhere. Tomorrow’s machine schedule isn’t done yet and the morning crew will need to know which job to start next, which orders are top priority, this week. The manager rubs his eyes, the lines on the spreadsheet are blurring. He grumbles to himself, “if only I had a crystal ball…”

It’s easy to picture the scene and empathize with the manager who has been forced to make decisions, with few tools or relevant, contextual facts to help him. He may feel like he’s being asked to conjure data out of thin air, magically seeing into the future to predict when inventory will be consumed, how and why. These are monumental questions with no easy answers. Yet, that is often what is expected of operations directors, shop floor supervisors and line of business managers.

Operational managers are tasked with keeping tight control on inventory in order to control costs. At the same time, however, they are also responsible for meeting customer expectations for on time delivery. A ready supply of raw materials and consumables must be on hand, or close at hand. Stock-outs and shortfalls can be disastrous, especially with mission critical parts or in industries with perishable goods.

Why managing inventory counts

Margins are tight in manufacturing today and there is constant pressure to reduce working capital. Controlling inventory is, as a result, a critical part of managing resources, controlling waste and keeping cash flowing. No CFO wants to hear that funds are tied up in costly, low turn spare parts that sit in inventory for months on end as the risk of obsolescence grows. As “just-in-time” inventory has become the mantra of lean practitioners, the days of amassing “just in case” safety stock has largely subsided.  But, then, as so often happens, the pendulum swings in the opposite direction. Manufacturers, looking to cut costs, are caught with gaps in their inventory and extended lead times as their global supply chain partners minimize inventories and utilize the lowest cost methods of delivery.

Modern, advanced analytics can prevent these major issues, without the aid of a crystal ball or fairy dust.  Analytics integrated with modern ERP solutions help manufacturers predict needs and derive meaningful insight from trends — past and future. Managers can go past the surface-level numbers and delve deeper into driving forces, tying usage to specific activities. The reasons behind the numbers allow managers to make fact-based decisions contingent on many non-obvious relationships and dependencies.

Today, knowing the reason behind an expense is far more important than knowing how much is spent. So, systems are moving away from simple financial reporting to now focus on providing valuable direction on managing a profitable, customer-centric business.  Predictive analytics allow the system to tie usage to activity, headcount, facility or truly any measured business variable.

Putting insight to use

When data is tied to multiple influencing factors and can be drilled into to find relationships, a manager can dig into the questions that used be dismissed as unanswerable mysteries, such as unusual spikes in product use, precipitous drops in the output of one production line, unprecedented shortfalls in one component or rapid depletions of a seemingly innocuous item. Such exceptions — at one time — may have been ignored or chalked up to the “cost of doing business.” Today, these details all merit further research in an effort to cut waste and identify intentional misuse or error. A conscientious manager will try to find causes to these exceptions, then, determine if the issue can be remedied, managed or forecasted.

Predicting needs is one of the most important benefits of advanced analytics. Trends can be analyzed and correlated into forecasts. The forecasting of sales and consumption of raw materials and consumables allow procurement managers to make long-term contract plans and to be more strategic in their purchasing policies. They can balance cost and availability with confidence. Just how reliable is this new vendor and what risk is associated with turning to an untested vendor in another hemisphere? With reliable demand forecasts, the purchasing department can accurately determine risk and make well informed decisions.

Predictive abilities also help manage the physical space and logistics of the inventory system. This allows warehouse managers to optimize material handling and storage systems. The inventory system can also be set up to minimize the amount of transfers between warehouse facilities, since handing of goods risks damage and adds to cost of storage.

The goal is to prevent surprises and be better prepared for fluctuations in sales, changing customer demands and evolving global conditions. A manufacturer who is prepared — with adequate inventory levels — will be able to build positive relationships with customers and optimize the use of resources.

By: Larry Korak, Director of Industry & Strategy, Industrial Manufacturing, Infor
Originally published at www.mbtmag.com

Leave a Reply