The Growing Number of “Warehouse Optimization System (WOS)” Providers Raises Questions About Established Warehouse Management System (WMS) Effectiveness

Established Warehouse Management System (WMS) providers are doing very well. For many of the major providers, including Blue Yonder, Infor, Tecsys, and Manhattan Associates, annual revenue has continued to grow at double-digit rates. For 2024, Blue Yonder reported a 14% Year-over-Year (YoY) increase in Software-as-a-Service (SaaS) revenue with WMS revenue growing 32%, and both Tecsys and Manhattan Associates saw total revenue increases of over 12% YoY.

While these WMS providers continue to expand their industry presence and improve in functionality, companies offering WMS add-ons and ways to improve existing solutions continue to pop up. AutoScheduler, Synkrato, and WareBee all provide a similar type of “Warehouse Optimization System (WOS),” a term recently coined by Synkrato, and are strictly focused on tackling warehouse planning challenges and improving system capabilities.

Key factors in the emergence of these solutions include issues with legacy WMSs that have unscalable architecture and are difficult to customize, as well as the ongoing issue of companies wrestling with data silos that continue to limit insight, analytics, and execution capabilities.

There are certain nuances differentiating the providers. AutoScheduler’s platform focuses on labor and dock optimization; Synkrato has placed Generative Artificial Intelligence (Gen AI) at the forefront, providing ways to engage with warehouse data in a more intuitive way; and WareBee provides a digital twin of current operations, with the ability to run simulations and provide recommendations. But while they differ in their unique selling points, they all operate in a relatively similar way, pooling operational data into a central point to provide more coordinated planning capabilities.

An important point to make is that these systems are not aiming to replace WMSs and WESs or offer the same type of service. They are intended to act as an additional, separate layer, focused on improving the current setup of a warehouse and allowing users to understand how changes will impact day-to-day picking operations. Similar providers like LogistiVIEW and Onomatic sit more on the Warehouse Execution System/Warehouse Control System (WES/WCS) side, focusing on orchestration, rather than the planning stage. It’s important to be aware of the differences when most providers market themselves in a very similar way.

It would be fair to think that this is something that a WMS should be providing, given that WMSs are the planning tool of the warehouse. But WMSs are predominantly only focused on the day-to-day and have yet to develop to the point of providing users with advanced planning tools via digital twins, simulations, and system-generated recommendations.

The ability to run simulations of operations based on specific changes, like those offered by WareBee, is a big selling point for the current state of warehousing. For example, when a company is considering implementing an Automated Storage & Retrieval System (ASRS) or an Autonomous Mobile Robot (AMR)-based system, it needs to understand if it will improve fulfillment, what Return on Investment (ROI) it will deliver as a result, and which goods (whether fast or slow moving) are best to put into the system. Justifying ROI is a major hurdle in adopting automation, so leveraging platforms like this can make a major difference.

As discussed in a recent ABI Insight, “Private Wireless Networks Should Play a Central Role in Supporting Warehouse Execution Software Challenges,” major WMS providers, including Blue Yonder, Korber, Manhattan Associates, and Infor, continue to take more of a horizontal, network-based approach to their offerings. The main goal is to connect the multiple facets of supply chain planning via a cloud-native, scalable platform, breaking down the many silos that exist across WMS, Transportation Management System (TMS), and Yard Management System (YMS) solutions and beyond. While this hasn’t necessarily come at the expense of improving established WMS platforms, the rise of WOS vendors does suggest that a hole has been left in the WMS planning sphere. WMS providers are and will continue to develop in this area, offering more tools for users to review, simulate, and optimize their operations, but the dedicated providers discussed above are moving a lot faster.

One issue that WOS vendors will face is the “system fatigue” that many companies often reference. Digitalizing more areas means taking on multiple systems from multiple providers and companies often struggle with the mismatch in capabilities and data silos that this creates. But in this case, having an entirely separate system may be the best approach. If only one planning system is used, there is only one opinion and limited inputs deciding what the best layout and workflows are. Particularly as systems are customized to unique workflows, it is very difficult for the same system to then take a step back and identify a new way of working. By adding an additional layer that can pool data from each system, more structural changes with greater efficiencies can be found.

Standard AI is well used for data analysis purposes, and Gen AI is beginning to be used as a way of interacting with warehouse data to provide insights in a faster, simpler way. But Agentic AI is really the next evolution of these types of platforms, with many of the vendors planning to announce new capabilities in this area through 2025. ABI Research expects these types of systems to be the best use cases for Agentic AI, given that they are more value-add than they are system-critical. Trusting AI at the execution level is still a challenge, but allowing a system to autonomously run simulations and provide the user with custom recommendations is a use that most people can get behind and will be able to add significant value in facilitating improvements all the way from basic workflow changes up to planning and integrating autonomous material handling capabilities.