Artificial Intelligence for Service Parts Planning

New Product Introduction

When you introduce a new product to the market, service parts must be available to support repair in the unfortunate event of a failure. For spare parts organizations, this concept encapsulates the crux of what they strive to achieve, yet many struggle to support products during the NPI phase.

With a new product, it’s hard to make projections for how it will perform. You try to pull together data and make an intelligent guess, but the lack of precise tools that can evaluate large data sets makes this suboptimal. For many companies, the NPI values are simply a guess. Customers either got their forecast too low, investing in fewer parts than they needed, or did the opposite, accumulating excess inventory.

The first instance leads to poor customer experience. Customers buy a new product, it breaks, they cannot get the spare parts they need for it from you, and downtime tarnishes the reputations of both you and your customer. Conversely, out of fear of falling short for your customers, you over invest in inventory to support the new product. You’ve wasted resources to purchase unnecessary parts and now you need to figure out what to do with them, accumulating even more expenses in the form of storage, recycling, or disposal costs. Not only does this scenario drain your resources, it generates unnecessary waste and carbon emissions, harming the environment.

Last-Time Buy Decisions

Challenges emerge when a product reaches the end of its road as well. As products approach the end of their lifecycle, it’s essential to accurately forecast the decline in parts demand and begin tapering off new part purchases where repaired parts can meet the service requirements throughout the end of life. In this case, you need to adjust your buying pattern to invest in just the right amount of inventory to see these products through to their end. Your customers will still rely on sunsetting products at different rates for varying amounts of time.

If you think of your product timeline as a tree trunk, your customers’ connections to the product are like branches. You must consider these as well. When you abruptly stop investing in all the accessories that support these products, any branch connected to the product will suffer and you’ll end up letting customers down. But as is the case with NPIs, over-investing in parts to service a terminal product leads to the accumulation of obsolete materials and wasted resources. To make matters more complicated, many of you have to make LTB decisions for multiple products at once. Some of them share replaceable materials, some of them don’t. Now you’re no longer just dealing with a tree, you have a forest with interlocking branches and roots to consider.

Additional considerations grow in criticality if you must make LTB decisions when a supplier advises that they’re no longer going to manufacture a part. As a result, planners must purchase a quantity of parts to support the products throughout the remainder of the lifecycle or as contracts warrant. This becomes even more complex as parts exist on multiple product bills of materials (BOM) in different lifecycle stages.

Baxter Planning’s Predictive Service Supply Chain Solution

Our customers have cited NPIs and LTBs as vulnerable for their businesses because the necessary technology to support these decisions simply was not ready.

These organizations had to rely on guesswork and excel spreadsheets, which only take them so far. This is why we’ve been working diligently to come through with the targeted support they need. Now that we have access to the technology necessary mixed with the needed domain expertise, we can solve transition-related business issues and we’re excited for them to see serious results.

The Power of BaxterPredict Prophet.ai

To target the complexity of NPI and LTB decisions, we use AI/ML to more accurately project product lifecycles through end of life.

After a few months of training, Prophet.ai allows Baxter Planning to project:

● The product lifecycle after as a little as 5 data points that may be simply your sales projection
● With up to an unprecedented 95% accuracy

Thanks to this, we estimate our customers will save 12% of their inventory spend on an annual basis. So, if you spend $20 million on service inventory on an annual basis, you can reduce that by about $2.5 million.

Prophet.ai also allows customers to identify excess inventory earlier than was ever possible before, enabling them to make more efficient use of these resources before they go to waste.

New Product Introductions with AI

Even with minimal initial data, the forecast provides accurate peak and end-of-production dates. As installed base data becomes available, the forecast adapts, improving in accuracy.

From there, Prophet.at drills down into the material level forecast with information from the product’s BOM. This approach allows us to see the material’s installed base forecast, interact with product life curves, and seamlessly integrate with the supply plan and demand projections.

Additionally, during implementation, the Baxter Planning team collaborates with customers to cleanse their product installed base. The ML model generates clusters, matching historical data for accurate lifecycle projections. This process extends to material lifecycle projections, considering failure rates and using precise product lifecycle projections. The integration allows for a comprehensive material installed base forecast, impacting the supply plan and driving demand projections, providing insights into potential demand spikes and their duration. With visibility into peaks of demand, customers will not feel as compelled to buy in excess early on.

With Prophet.ai, we now can also discern when needs for certain materials will spike when a new product is introduced mid-lifecycle. Traditional statistical forecasts would miss this increase in demand, not considering the needs of the new product, and show an inaccurate downward trend in material demand. Now our customers can see this demand bump ahead of time and prepare for it without overcompensating.

Last-Time Buys with AI

Prophet.ai leverages Machine Learning to generate several scenarios across the spectrum from best case to worst case. In turn, they can make smarter decisions when it comes to purchasing material to support terminal products.

The flexibility of Prophet.ai also goes beyond individual product considerations. Customers can benefit from the platform’s advanced capabilities by grouping LTB decisions based on shared materials and components.

Supply Chain Sustainability

Your product ecosystem is connected to an even broader environmental ecosystem, and we designed Prophet.ai with this in mind. Not only do our customers constantly seek ways to firm their footing on a turbulent business landscape, but they also care about how their carbon footprint impacts the ever-changing climate. In an era where environmental sustainability is a critical business imperative, we know our customers face mounting pressure to play a responsible role in the climate crisis.

According to a recent survey distributed by PRNewswire, 80% of consumers prioritize environmentally responsible practices when engaging with brands. Regulatory frameworks worldwide are also tightening, with an increasing number of jurisdictions implementing stringent environmental standards.

By accurately predicting product lifecycles and spare parts demand, you can avoid overstocking, minimizing unnecessary waste and carbon emissions associated with surplus inventory. This aligns with supply chain sustainability goals and also reduces the carbon footprint associated with the production, storage, and disposal of excess materials.

In the context of NPIs, preventing both underinvestment and overinvestment in inventory becomes key to curbing environmental harm. Overinvesting can lead to resource-intensive storage, recycling, or disposal processes; underinvesting leads to scenarios where you must expedite missing parts next-flight out or similar, increasing carbon emissions.

Prophet.ai promotes sustainability-focused decisions in the LTB phase as well. By grouping LTB decisions for shared materials and components, businesses can optimize their purchasing strategies, ensuring more efficient use of resources through the end of a product’s lifecycle. This minimizes the accumulation of obsolete materials and aligns with supply chain sustainability goals by reducing waste and promoting responsible resource management.

Get Started with BaxterPredict Prophet.ai

Many Service Supply Chain professionals deal with a diverse range of products across sprawling global networks. Several of our customers have identified LTB and NPI scenarios as some of the weakest links in their operations.

In the Service Supply Chain, the challenges of determining optimal inventory levels are compounded when factoring in the uncertainties associated with changing product-support demands. Historically, companies have grappled with inaccurate predictions, resulting in either underinvestment, leading to poor customer experiences, or over investment, contributing to excess inventory and resource wastage.

Prophet.ai addresses these longstanding issues by leveraging ML and vast datasets to predict product lifecycles and spare parts demand with unprecedented accuracy.

The benefits of Prophet.ai extend beyond mere accuracy in predictions. The system empowers service organizations to:

• Save an estimated 12% on their annual inventory spend
• Enhance customer satisfaction and thusly Net Promoter Score
• Safeguard brand integrity by ensuring an adequate supply of spare parts for new products.

The significance of this technology is magnified when considering its positive environmental impact, aligning with the growing imperative for sustainability in 2024.

With this powerful technology now at our fingertips, we’re excited to watch our customers build more sustainable businesses from the inside out.

Ready to get started?