Will the Evolution of Machine Learning Systems Lead to Self- Thinking Supply Chains?

By Henry Canitz, Director Product Marketing & Business Development, Logility

content-image

Henry Canitz, Director Product Marketing & Business Development, Logility

The use of Artificial Intelligence (AI) in managing the supply chain is inevitable. In fact, today there are early examples of the potential of AI to improve supply chain efficiencies and provide better and more optimized supply chain decisions.

Machine learning is a type of artificial intelligence that gives computers the ability to learn without explicit programming. Computer programs that use machine learning, improve their ability to solve problems as exposed to new data. The latest Gartner Supply Chain Hype Cycle, published in July 2016, shows machine learning approaching the “Peak of Inflated Expectations.” Gartner predicts that mainstream adoption of Machine Learning is at least five years away, potentially ten. Still, typing “Machine Learning” and “Supply Chain Planning” into Google delivers more than 42,200 results, a topic which supply chain management professionals are thinking, talking, and writing about.

In reality, the supply chain planning mindshare spent on machine learning is a fraction of that spent on reducing costs, improving customer service and driving new revenue. Type “Cost Savings” and “Supply Chain Management” into Google, and you get 450,000 results. One could argue that machine learning could contribute to meeting cost, revenue and customer service goals. However, there is still more focus today on supply chain management basics like forecasting, inventory management, and Sales & Operations Planning (S&OP) than AI driven automated operations.

Still, there is a general buzz in the field of supply chain management regarding the ability of artificial intelligence and machine learning to substantially transform the supply chain of the future. In a recent survey of CEOs conducted by Gartner, over 48 percent expect their supply chain to be unrecognizable by 2020. There is a general belief amongst senior supply chain management that supply chains of the future need to be quicker and smarter with capabilities that automate the routine and streamline the ability to pursue opportunities and mitigate risks.

The critical question is, “How do we get to a highly automated supply chain from where most companies are today?”

Most pundits believe that an essential building block to moving up the automation maturity curve is the establishment of integrated supply chain operations enabled by a Supply Chain Planning (SCP) System of Record (SOR). Most companies of any size consider an Enterprise Resource Planning (ERP) system to be an essential tool for managing their business. An ERP is the SOR for financial, customer, product, and other data used to run the business. Unfortunately, many of these same companies are yet to implement an SCP SOR preferring to operate a critical part of their business on isolated point solutions, inadequate legacy solutions, and spreadsheets. The SCP SOR provides a single place to manage demand, inventory, replenishment, order promising, production, and manufacturing scheduling.

Optimization driven by algorithmic planning is an early form of machine learning that relies on a set of provided information to automatically make optimal recommendations


Gartner found that companies need an SCP SOR to provide a strong foundational planning layer, to build more differentiated and innovative capabilities like automated operations and machine learning capabilities.

Once a company develops integrated operating capabilities supported by an SCP SOR, some believe the path to AI and Machine Learning is through climbing the levels of analytics maturity. Gartner has proposed five stages of analytics maturity:

1. Descriptive – Determining what happened through reports, KPIs, etc.

2. Diagnostic – Discovering why it happened through Root Cause Analysis, 6-Sigma Tools, etc.

3. Predictive – Understanding what is likely to happen through forecasts, what-if scenarios, etc.

4. Prescriptive–Determining the optimal response through algorithmic optimization including automating this response.

5. Cognitive–Developing new insights or recommended actions based on artificial intelligence and machine learning.

What capabilities might you investigate to climb the ladder of analytics maturity and start your journey towards transforming to a more automated and optimized supply chain? The logical place to start is to build algorithmic optimization capabilities to continually analyze the state of your supply chain and recommend or automatically execute plans to meet customer requirements. Optimization driven by algorithmic planning is an early form of machine learning that relies on a set of provided information (supply chain facilities and capacities, transportation lanes and capacities, customer service requirements, profit requirements, etc.) to automatically make optimal recommendations.

One powerful example of algorithmic optimization is Multi-Echelon Inventory Optimization (MEIO) used to analyze and optimize inventory positions. The current inventory-to-sales ratio in the United States sits at 1.38 (See Figure 1). One cause of this glut of inventory is the emergence of omni channel retailing. The “Amazon effect” of free and fast shipping, easy returns, and everyday low prices has changed customer expectations, and the way products get into the hands of the consumer. It’s a struggle to keep up with the mind-bending rate of change that prevents companies from having the right inventory positioned at the right location to service customers. Companies that respond by increasing buffer inventory by channel, based on outdated information fail to attack the underlying issue of stock-outs, a fundamental shift in fulfillment. Multi-echelon Inventory Optimization automatically seeks the optimal balance of inventory at the right locations and provides inventory parameters and positions by stocking location to establish optimal buffer locations and quantities. Embracing MEIO can reduce total inventories by upwards of 30 percent while maintaining or improving customer fill rates. MEIO is an example of basic Machine Learning that is available today.

Attaining the full benefits of machine learning will be an evolutionary process. We must build a strong foundation of integrated supply chain planning capabilities first, and then work to move up the analytics maturity curve. The introduction of machine learning into most supply chain organizations will take years, but that shouldn’t stop supply chain professionals from planning for the future or taking advantage of some of the basic machine learning solutions available today. Implementing algorithmic optimization technologies today builds the kind of expertise and experience that will ease the adoption of advanced artificial intelligence solutions in the future.

Headquartered in Atlanta, Logility is a provider of collaborative supply chain optimization and advanced retail planning solutions that helps small and large enterprises, and Fortune 500 companies.

Read Also

Augmenting Supply Chain with Real -Time Tracking and Tracing Systems

Augmenting Supply Chain with Real -Time Tracking and Tracing Systems
Greg Toornman, Vice-President, Global Materials, Logistics, and Freight Management, Agco Corporation
Autonomous Technology Leans into Logistics

Autonomous Technology Leans into Logistics
Diane Sacra, Director of Marketing, Universal Logistics Holdings, Inc.