Growing Complexity
Growing ComplexityThe complexity of running the warehouse only continues to increase. Supply chain leaders face macro-challenges such as the pressure for sustainability, labor shortages and the effects of inflation on operating margins.
Layer on the daily micro disruptions that every warehouse experiences regardless of size – unexpected or delayed deliveries, inventory shortages, quality issues and scheduled labor or equipment mismatches versus real-time needs—and suddenly running the warehouse becomes unmanageable.
Backwards Solutions
Logisticians and warehouse operators have worked to rise to the challenge; however, there is only so much complexity that a warehouse team can effectively manage using many of today’s siloed, rules- and history-based warehouse management solutions. Too often, the challenge is met by throwing added resources at problems to close the gap. And unfortunately, those resources may not actually add much value.
For example, slotting and picking usually consume more than half of warehouse labor costs. Studies estimate that a typical warehouse picker spends over 50% of their shift just in travel time winding their way across the warehouse to or from a product pick. Warehouses also struggle with being over or understaffed and rarely strike the balance of what is “just right” for the day’s staffing needs.
This is because the data that the warehouse is working on includes historical receipts, demand, picks and shipments. When it does look into the future, it too often uses a forecast or plan that can be weeks, if not months, out of date.
AI Helps Close the Complexity Gap
Acknowledging that warehouse complexity is not going away, what can be done?
We envision AI and Machine Learning as the path to close the complexity gap and move the warehouse to the next level of efficiency and effectiveness.
AI and Machine Learning can take signals from historical data, blend them with real-time updates, future forecasts, predictive learning models and even include signals from other connected solutions. Together, these signals provide the warehouse team with recommendations that can optimize activity across the warehouse for today and into the future.
AI and Machine Learning can be the partner that the warehouse team has been looking for to simplify their complexity. We can already see ways that this can transform warehouse operations.
Agentic AI
Every warehouse team has that one person who understands the warehouse and brings an unrivaled depth of experience that helps the warehouse run.
Now, take that knowledge and expertise and imagine it on an interactive device and in your team’s hands 24/7/365.
Agentic AI can incorporate all those signals and provide recommendations in real-time to your warehouse team on the most effective use of their time. What trailer needs to be worked next and why? What is the best use of your slotting team’s time to reduce pick travel time and improve efficiency?
All while providing your entire team with clear, understandable explanations of why the recommendation was made and how it can help your team and operations.
AI in the Yard
AI and Machine Learning can help to reduce complexity in the warehouse yard.
Today, trailers are often prioritized for unloading mainly to support facility service levels and to avoid detention charges.
With AI and Machine Learning, Warehouse Management can utilize real-time signals to “see” the contents of each trailer, the status of pending orders against those contents and their priority against every other trailer in the yard.
AI and Machine Learning can provide real-time recommendations based on available scheduled resources, select the highest priority trailers to be worked next and anticipate the best dock door to most efficiently slot the product or cross dock for immediate shipment.
All to reduce complexity in the yard and the receiving dock.
AI Within the Facility
Slotting and picking are among the most labor-intensive activities within the facility. Most slotting decisions are made literally in the moment based on available slots. The results are products distributed across the warehouse based on thousands of disconnected slotting decisions.
With the introduction of AI and Machine Learning to the process, warehouse management can analyze each individual slotting decision, including signals such as history, volume estimates, pick affinity, optimal task and travel time, as well as available locations. Then, balance them against configured guardrails such as efficiency and movement costs to determine slotting locations that continuously optimize the warehouse for tomorrow’s needs, instead of just today’s availability.
The result is reduced complexity, which decreases pick travel time and increases pick efficiency since the slotting decisions have been continuously optimized over time to place products for faster picking.
AI and Resource Management
Accurately forecasting and then managing warehouse resources in real-time is a significant part of a warehouse leader’s work.
In today’s warehouse, resource forecasts are typically based on a combination of historical data and demand forecasts that usually do not accurately reflect what is needed in real-time. So, resource forecasting too often results in the warehouse being chronically over or understaffed.
When forecasting resources with AI and ML, solutions can incorporate historical data but also the latest real-time forecasted data to more accurately project resource needs days or months in advance, improving staffing and scheduling at a much more granular level.
But even the most intelligent resource forecast needs to be adjusted when today’s emergency intrudes. With AI-powered visibility, the warehouse management solution can ingest all these various signals, see the priority of workflows across the entire building and then shift available resources in real time from lower-priority work to mitigate today’s emergency.
The warehouse management solution is continuously updated and can re-prioritize tasks in real time, ensuring that all work across the facility stays on track and reducing the complex firefighting work that warehouses complete daily.
All these recommendations can be reviewed and approved by the warehouse team based on clear and understandable explanations from the solution’s AI.
Conclusion
All of this sounds like science fiction. But it is all very real.
Software providers are seeing more logistics and supply chain clients that need and inquire about AI and Machine Learning capabilities as part of their warehouse management or other solution sets.
To address the growing complexity and macro- and micro-disruptions, the warehouse team needs a trusted partner that can learn, improve and understand their facility and its capabilities.
A trusted partner that can help to reduce complexity and keep the facility working efficiently, not just today but every day.
AI and Machine Learning is ready to be that partner.
Steve Ross is part of Blue Yonder’s Solution and Industry Marketing team focusing on Supply Chain Execution.
With almost 30 years of operations, logistics, and e-commerce experience. Steve is an advocate for how technology can help free up the warehouse and store teams from the obstacles that legacy processes, solutions, and thinking have put in their path.
Steve believes in that Blue Yonder’s industry-leading supply chain platform and cognitive capabilities can be the partner the warehouse team needs to simplify their workload and improve efficiency.
Before joining Blue Yonder, Steve held multiple leadership roles, serving as the “translator” between the Logistics, Operations, Planning, and Buying teams. Steve has led various omnichannel implementations and has a background in Lean Six Sigma and project management.
Steve has an MBA from Roosevelt University, and a BA from the University of Arizona.
In his spare time, you will often find him in his kitchen baking sourdough bread or something sweet.
