• OneTrack Marketing

How Computer Vision can reduce forklift travel

We often hear about initiatives at logistics and manufacturing companies to reduce travel distance in their material handling processes. Common areas of focus are:

  • Inbound put-away

  • Finished goods to storage

  • Outbound staging

When speaking with Process and Industrial Engineers, it appears that the concept of travel distance is often considered closely correlated with throughput. In an ideal world, reducing forklift travel should directly increase productivity since less distance has to be covered during every move.


After spending years building OneTrack's proprietary indoor-localization platform OneTrack LX and analyzing millions of data points from our fleet of real-time AI camera sensors, we have come to the conclusion that this is only partially true.


While long-distance travel is certainly a factor in warehouses, it is often a much smaller actionable improvement opportunity than expected. To make meaningful changes to overall travel distance based on long-distance travel routes, most warehouses would have to make massive layout adjustments every day.


The reality is, that the true area of opportunity for immediate improvement is related to a metric we refer to as "travel distance density". To understand why this is the case, consider the following travel path, as measured by our camera-based indoor localization system:

This is the travel path of an operator completing a list pick of several cases in a far-away location relative to the starting dock location (DOCK121C) where the finished pallet is also being staged.


We can see that the operator makes several stops along the way to obtain cases. If we wanted to reduce the long-distance travel for this operator, we would have to adjust the drop-off locations of the cases during inbound putaway or select a different dock door for the trailer. This particular pallet of cases that the operator is obtaining, however, is only one of 20 pallets that are going into this trailer and the dock location might very well be the best location based on all 20 pallets that are being loaded. The reality is that there are a large number of constraints on where products can be stored in the warehouse and travel distance for one particular assignment is only one of many of these factors. Furthermore, those constraints could change on a daily basis requiring costly WMS reconfiguration.


In order to actually increase operator productivity, the critical metric to look at is travel distance density because it can be used for coaching, even when the operator is taking the most efficient route possible.


We define travel distance density as the total distance traveled near transaction start and end locations. In the example above, travel distance density is high every time the operator starts or stops to obtain or place cases and it is highest as the operator navigates the dock door area to stage the finished pallet.


Other than extreme outliers, the majority of distance traveled is usually covered while forklifts are navigating near or around the start/stop locations, completing one of the following activities:

  • Stacking pallets

  • Realigning pallets

  • Rotating pallets

  • Obtaining pallets

  • Placing pallets

These activities are, incidentally, also the ones where the most time passes and ultimately what defines the productivity bottleneck for any employee in any process. The time spent traveling to and from far away locations is usually minimal compared to the overall time spent on a task. This is critical because, in the end, employees are compensated based on hours worked, not distance traveled.


Using OneTrack's Artificial Intelligence enabled labor management solution, it is possible to automatically identify these types of travel distance density outliers and define workflows that focus on activity-specific coaching. This increases operator productivity and reduces the labor cost per unit.


If you want to learn more about measuring and optimizing forklift travel in warehouses, book a demo!