Are you confused about what is AI and what is machine learning? Are you unsure why knowing more will help you with your job in inventory planning? Don’t despair. You’ll be ok, and we’ll show you how some of whatever-it-is can be useful.

What is and what isn’t

What is AI and how does it differ from ML? Well, what does anybody do these days when they want to know something? They Google it. And when they do, the confusion starts.

One source says that the neural net methodology called deep learning is a subset of machine learning, which is a subset of AI. But another source says that deep learning is already a part of AI because it sort of mimics the way the human mind works, while machine learning doesn’t try to do that.

One source says there are two types of machine learning: supervised and unsupervised. Another says there are four: supervised, unsupervised, semi-supervised and reinforcement.

Some say reinforcement learning is machine learning; others call it AI.

Some of us traditionalists call a lot of it “statistics”, though not all of it is.

In the naming of methods, there is a lot of room for both emotion and salesmanship. If a software vendor thinks you want to hear the phrase “AI”, they may well say it for you just to make you happy.

Better to focus on what comes out at the end

You can avoid some confusing hype if you focus on the end result you get from some analytic technology, regardless of its label. There are several analytical tasks that are relevant to inventory planners and demand planners. These include clustering, anomaly detection, regime change detection, and regression analysis. All four methods are usually, but not always, classified as machine learning methods. But their algorithms can come straight out of classical statistics.

Clustering

Clustering means grouping together things that are similar and distancing them from things that are dissimilar. Sometimes clustering is easy: to separate your customers geographically, simply sort them by state or sales region. When the problem is not so dead obvious, you can use data and clustering algorithms to get the job done automatically even when dealing with massive datasets.

For example, Figure 1 illustrates a cluster of “demand profiles”, which in this case divides all a customer’s items into nine clusters based on the shape of their cumulative demand curves. Cluster 1.1 in the top left contains items whose demand has been petering out, while Cluster 3.1 in the bottom left contains items whose demand has accelerated.  Clustering can also be done on suppliers. The choice of number of clusters is typically left to user judgement, but ML can guide that choice.  For example, a user might instruct the software to “break my parts into 4 clusters” but using ML may reveal that there are really 6 distinct clusters the user should analyze. 

Figure 1: Clustering items based on the shapes of their cumulative demand

Anomaly Detection

Demand forecasting is traditionally done using time series extrapolation. For instance, simple exponential smoothing works to find the “middle” of the demand distribution at any time and project that level forward. However, if there has been a sudden, one-time jump up or down in demand in the recent past, that anomalous value can have a significant but unwelcome effect on the near-term forecast.  Just as serious for inventory planning, the anomaly can have an outsized effect on the estimate of demand variability, which goes directly to the calculation of safety stock requirements.

Planners may prefer to find and remove such anomalies (and maybe do offline follow-up to find out the reason for the weirdness). But nobody with a big job to do will want to visually scan thousands of demand plots to spot outliers, expunge them from the demand history, then recalculate everything. Human intelligence could do that, but human patience would soon fail. Anomaly detection algorithms could do the work automatically using relatively straightforward statistical methods. You could call this “artificial intelligence” if you wish.

Regime Change Detection

Regime change detection is like the big brother of anomaly detection. Regime change is a sustained, rather than temporary, shift in one or more aspects of the character of a time series. While anomaly detection usually focuses on sudden shifts in mean demand, regime change could involve shifts in other features of the demand, such as its volatility or its distributional shape.  

Figure 2 illustrates an extreme example of regime change. The bottom dropped out of demand for this item around day 120. Inventory control policies and demand forecasts based on the older data would be wildly off base at the end of the demand history.

Figure 2: An example of extreme regime change in an item with intermittent demand

Here too, statistical algorithms can be developed to solve this problem, and it would be fair play to call them “machine learning” or “artificial intelligence” if so motivated.  Using ML or AI to identify regime changes in demand history enables demand planning software to automatically use only the relevant history when forecasting instead of having to manually pick the amount of history to introduce to the model. 

Regression analysis

Regression analysis relates one variable to another through an equation. For example, sales of window frames in one month may be predicted from building permits issued a few months earlier. Regression analysis has been considered a part of statistics for over a century, but we can say it is “machine learning” since an algorithm works out the precise way to convert knowledge of one variable into a prediction of the value of another.

Summary

It is reasonable to be interested in what’s going on in the areas of machine learning and artificial intelligence. While the attention given to ChatGPT and its competitors is interesting, it is not relevant to the numerical side of demand planning or inventory management. The numerical aspects of ML and AI are potentially relevant, but you should try to see through the cloud of hype surrounding these methods and focus on what they can do.  If you can get the job done with classical statistical methods, you might just do that, then exercise your option to stick the ML label on anything that moves.

Did you know that it was Benjamin Franklin who invented the lightning rod to protect buildings from lightning strikes? Now, it’s not every day that we must worry about lightning strikes, but in today’s unpredictable business climate, we do have to worry about supply chain disruptions, long lead times, rising interest rates, and volatile demand. With all these challenges, it’s never been more vital for organizations to accurately forecast parts usage, stocking levels, and to optimize replenishment policies such as reorder points, safety stocks, and order quantities.  In this blog, we’ll explore how companies can leverage innovative solutions like inventory optimization and parts forecasting software that utilize machine learning algorithms, probabilistic forecasting, and analytics to stay ahead of the curve and protect their supply chains from unexpected shocks.

Spare Parts Planning Solutions
Spare parts optimization is a key aspect of supply chain management for many industries. It involves managing the inventory of spare parts to ensure they are available when needed without having excess inventory that can tie up capital and space. Optimizing spare parts inventory is a complex process that requires a deep understanding of usage patterns, supplier lead times, and the criticality of each part for the business.

In this blog, our primary emphasis will be on the crucial aspect of inventory optimization and demand forecasting. However, other approaches highlighted below for spare parts optimization, such as predictive maintenance and 3D printing, Master Data Management, and collaborative planning should be investigated and deployed as appropriate.

1. Predictive Maintenance: Using predictive analytics to anticipate when a part is likely to fail and proactively replace it, rather than waiting for it to break down. This approach can help companies reduce downtime and maintenance costs, as well as improve overall equipment effectiveness.
2. 3D printing: Advancements in 3D printing technology are enabling companies to produce spare parts on demand, reducing the need for excess inventory. This not only saves space and reduces costs but also ensures that parts are available when needed.
3. Master Data Management: Data Management platforms ensure that part data is properly identified, cataloged, cleansed, and organized. All too often, MRO organizations hold the same part number under different SKUs. These duplicate parts serve the same purpose but require different SKU numbers to ensure regulatory compliance or security.  For example, a part used to support a government contract may be required be sourced from a US manufacturer to stay in compliance with “Buy America” regulations.  It’s critical that these part numbers be identified and consolidated into one SKU, when possible, to keep inventory investments in check.
4. Collaborative Planning: Collaborating with suppliers and customers to share data, forecasts, and plan demand can help companies reduce lead times, improve accuracy, and reduce inventory levels. Forecasting plays an essential role in collaboration as sharing insights on purchases, demand, and buying behavior ensures suppliers have the information they need to ensure stock availability for customers.

Inventory Optimization
Abraham Lincoln was once quoted as saying, “Give me six hours to chop down a tree, and I will spend the first four sharpening the axe”? Lincoln knew that preparation and optimization were key to success, just like organizations need to have the right tools, such as inventory optimization software, to optimize their supply chain and stay ahead in the market. With inventory optimization software, organizations can improve their forecasting accuracy, lower inventory costs, improve service levels, and reduce lead times. Lincoln knew that sharpening the axe was necessary to accomplish the job effectively without overexerting.  Inventory Optimization ensures that inventory dollars are allocated effectively across thousands of parts helping ensure service levels while minimizing excess stock.

Spare parts play a decisive role in maintaining operational efficiency, and the lack of critical parts can lead to downtime and reduced productivity. The sporadic nature of spare parts demand makes it difficult to predict when a specific part will be required, resulting in the risk of overstocking or understocking, both of which can incur costs for the organization.  Additionally, managing lead times for spare parts poses its own set of challenges. Some parts may have lengthy delivery times, necessitating the maintenance of adequate inventory levels to avoid shortages. However, carrying excess inventory can be costly, tying up capital and storage space.

Given the myriad of challenges facing materials management departments and spare parts planners, planning demand, stocking levels, and replenishment of spare parts without an effective inventory optimization solution is akin to attempting to chop down a tree with a very blunt axe! The sharper the axe, the better your organization will be able to contend with these challenges.

Smart Software’s Axe is the Sharpest
Smart Inventory Optimization and Demand Planning Software uses a unique empirical probabilistic forecasting approach that results in accurate forecasts of inventory requirements, even where demand is intermittent. Since nearly 90% of spare and service parts are intermittent, an accurate solution to handle this type of demand is required.   Smart’s solution was patented in 2001 and additional innovations were recently patented in May of 2023 (announcements coming soon!).  The solution was awarded as a finalist in the APICS Technological Innovation Category for its role in helping transform the resource management industry.

The Role of Intermittent Demand
Intermittent demand does not conform to a simple normal or bell-shaped distribution that makes it impossible to forecast accurately with traditional, smoothing-based forecasting methods.  Parts and items with intermittent demand – also known as lumpy, volatile, variable or unpredictable demand – have many zero or low-volume values interspersed with random spikes of demand that are often many times larger than the average. This problem is especially prevalent in companies that manage large inventories of service and spare parts in industries such as aviation, aerospace, power and water supply and utilities, automotive, heavy asset management, high tech, as well as in MRO (Maintenance, Repair, and Overhaul).

Scenario Analysis
Smart’s patented and award-winning technology rapidly generates tens of thousands of possible scenarios of future demand sequences and cumulative demand values over an item’s lead time. These scenarios are statistically similar to the item’s observed data, and they capture the relevant details of intermittent demand without relying on the assumptions commonly made about the nature of demand distributions by traditional forecasting methods. The result is a highly accurate forecast of the entire distribution of cumulative demand over an item’s lead time. The bottom line is that with the information these demand distributions provide, companies can easily plan safety stock and service level inventory requirements for thousands of intermittently demanded items with nearly 100% accuracy.

Benefits
Implementing innovative solutions from Smart Software such as SmartForecasts for statistical forecasting, Demand Planner for consensus parts planning, and Inventory Optimization for developing accurate replenishment drivers such as min/max and safety stock levels will provide forward-thinking executives and planners with better control over their organization’s operations.  It will result in the following benefits:

1. Improved Forecasting Accuracy: Accurate demand forecasting is fundamental for any organization that deals with spare parts inventory management. Inventory optimization software uses sophisticated algorithms to analyze historical usage patterns, identify trends and forecast future demand with a high degree of accuracy. With this level of precision in forecasting, organizations can avoid the risk of overstocking or understocking their spare parts inventory.
2. Lower Inventory Costs: One major challenge that supply chain leaders face when dealing with spare parts inventory management is the cost associated with maintaining an optimal stock of spares at all times. By optimizing inventory levels using modern technology systems like artificial intelligence (AI), machine learning (ML), and predictive analytics, organizations can reduce carrying costs while ensuring they have adequate stocks available when needed.
3. Improved Service Levels: When it comes to repair and maintenance services, time is money! Downtime due to the unavailability of critical spare parts can result in lost productivity and revenue for businesses across industries such as manufacturing plants, power generation facilities, or data centers managing IT infrastructure equipment. Optimizing your spare parts inventory ensures that you always have the right amount on hand, reducing downtime caused by waiting for deliveries from suppliers.
4. Reduced Lead Times: Another benefit that accrues from accurate demand forecasting through modern warehouse technologies is reduced lead time in delivery which leads to better customer satisfaction since customers will receive their orders faster than before thus improving brand loyalty. Therefore, the adoption of new strategies driven by AI/ML tools creates value within supply chain operations leading to increased efficiency gains not only limited reductionism cost but also streamlining processes related to production scheduling, logistics transportation planning among others

Conclusion
Through the utilization of inventory optimization and demand planning software, organizations can overcome various challenges such as supply chain disruptions, rising interest rates, and volatile demand. This enables them to reduce costs associated with excess storage space and obsolete inventory items. By leveraging sophisticated algorithms, inventory optimization software enhances forecasting accuracy, ensuring organizations can avoid overstocking or under-stocking their spare parts inventory. Additionally, it helps lower inventory costs by optimizing levels and leveraging technologies like artificial intelligence (AI), machine learning (ML), and predictive analytics. Improved service levels are achieved as organizations have the right quantity of spare parts readily available, reducing downtime caused by waiting for deliveries. Furthermore, accurate demand forecasting leads to reduced lead times, enhancing customer satisfaction and fostering brand loyalty. Adopting such strategies driven by AI/ML tools not only reduces costs but also streamlines processes, including production scheduling and logistics transportation planning, ultimately increasing efficiency gains within the supply chain.