Artificial Intelligence (AI) has become a major buzzword, with nearly every warehouse automation company claiming to incorporate some form of AI into their offerings. However, the term has become so overused and generic that it’s gradually losing its meaning. This not only misleads end customers but also makes it difficult for warehouse automation vendors to understand where AI is truly heading – and what genuine opportunities exist to deliver value.
This article explores the most valuable AI use cases for customers and offers guidance on how automation vendors should prioritize their AI development efforts.
What Is Artificial Intelligence?
The term “artificial intelligence” is often used loosely in the warehouse automation industry, so it’s important to clarify what we mean by AI in this context.
While there’s no single universally accepted definition, we refer to the one used by the National Artificial Intelligence Initiative and the White House Executive Order on AI, which defines AI as:
“A machine-based system that can, for a given set of human-defined objectives, make predictions, recommendations, or decisions influencing real or virtual environments.”
This broad definition can be broken down into three key subsets:
- Machine Learning (ML): A branch of AI and computer science that uses data and algorithms to mimic how humans learn, with models that improve over time.
- Deep Learning: A subfield of machine learning that uses layered neural networks (known as deep neural networks) to simulate complex human decision-making.
- Generative AI: A class of deep learning models that can generate high-quality content (e.g. text, images, code) based on the data they were trained on.
How to think about potential AI services
To identify and evaluate AI use cases, it’s helpful to view them across the warehouse automation value chain – from development and delivery to execution and maintenance. Because the potential applications for AI are vast, using a value-chain-based framework offers structure and clarity.
The table below summarizes the value chain for warehouse automation vendors and the corresponding AI use cases, which we’ll explore in more detail.
AI services across the value chain
However, just identifying use cases across the value chain doesn’t address a critical question: How should vendors prioritize their AI efforts?
To answer this, we need to evaluate which AI services offer either:
- The lowest barrier to entry for warehouse automation vendors
- The highest value, either in terms of customer benefits or internal efficiency gains (e.g. reduced software development time through AI copilots)
This is a key area of our ongoing research, and we expect to release more detailed insights soon. In the meantime, our initial findings are summarized in the matrix below, which maps barrier to entry vs. value added. The best “quick wins” are those AI services that combine high value with low implementation complexity.
Priority matrix for AI developments
Note: Analysis is based on interview insights, without quantitative analysis done. Barrier to entry is assessed specifically for warehouse automation vendors. The barrier may be higher or lower for different company types. Value is based on perceived value, based on qualitative research interviews, rather than a quantitative value analysis.
The need for a future-proof software foundation
Even with promising AI opportunities, warehouse automation vendors must first build an AI-ready software foundation. If there’s one takeaway from this article, it’s this: AI services can only thrive on a modern, flexible, API-first software platform.
Many vendors still operate on outdated, monolithic architectures, often the result of years of legacy development and acquisitions. These systems present major barriers to adopting AI in a meaningful way. To fully leverage AI’s benefits, vendors must adopt a modern, microservices-based architecture with API-first design.
In addition, software must support open interfaces and standard integrations to allow for seamless integration of machine learning operations (MLOps). Closed, proprietary systems make it nearly impossible to scale or maintain AI capabilities effectively.
The table below compares legacy systems with modern, AI-ready architectures:
What a modern software architecture should look like
Hybrid architectures: Balancing cloud and on-premise needs
It’s true that latency limitations often prevent full migration to the cloud – especially for real-time applications like warehouse control systems (WCS) – which limits the ability to adopt a true API-first architecture using micro-services. In such cases, vendors must develop a clear hybrid strategy; keeping latency-sensitive components on-premise, while migrating compute-heavy services to the cloud.
Leading vendors already operate in this way. For example, Knapp often deploys its core WCS platform, KiSoft, on-premise, but offers six complementary cloud-based services as part of its broader software suite:
- KiSoft Analytics
- RedPILOT
- KiSoft CMMS
- KiSoft Genomix
- KiSoft FCS
- KiSoft AI
This hybrid model enables flexibility, scalability, and faster AI deployment across the software stack.
Conclusion: Lay the right foundation first
While AI offers exciting opportunities in warehouse automation, success begins with the right technical foundation. Without modern, modular software, even the most promising AI ideas will be difficult to implement or scale.
By investing in future-proof software architecture, vendors position themselves to:
- Unlock AI use cases with real customer value
- Reduce internal development costs through automation
- Stay competitive as AI capabilities mature across the industry
In short, building an AI-ready platform is not optional, it’s a strategic imperative for any warehouse automation company aiming to be a leader of the next generation of intelligent automation.
About the Author
Rueben Scriven, Research Manager
Rueben is one of the world’s leading warehouse automation industry analysts. He’s spoken at all the leading industry events, and featured in the Financial Times, the WSJ, The Economist, Reuters, and CNBC, along with countless trade publications. He leads the Warehouse Automation research practice at Interact Analysis, a market intelligence firm focused on supply chain automation technologies. In recent years, his group’s research and analysis has expanded into warehouse software covering the whole tech-stack from sub-system control software, to execution and management software.
