We often measure productivity and capacity using units that are readily available. Sometimes this can be described as looking for lost car keys under a streetlamp not because that’s where they were lost but because that’s where the light is better. Examples of readily available units for a wall panel line are lineal feet and board feet. Because these metrics are not very reflective of the variety of wall panels produced, seasoned managers will use variations on a theme – categorizing different types of walls and/or dividing board feet by lineal feet to try to obtain some measure of relative complexity. If we look at a long enough period of time, we can apply factors to get these metrics to accurately predict average production rates.
Issues with the standard solutions – What is lost is variability of actual versus predicted times on a given shift, job, or specific panel. This means we don’t have a good grasp on whether the pricing model is using one type of panel or one type of job to subsidize another.
The AMT Robotics Solution – Why not use attributes that to this point are harder to get in order to better predict production times, especially at the bottleneck of the line (which sets the overall pace of the line)? Even better, why not match these attributes with the specific capabilities of the machine?
For example, we can divide the number of sheathing nails in a panel by the number of nailing heads to predict time. We can also factor in the density of the nailing pattern, nail reload time, and time to move the next panel into position (more predictable on a powered conveyor versus a manual process). Once we are confident that we can accurately predict the time to sheath an individual panel, then we can use sensors to dig into the various reasons why actual times are different than predicted times on an aggregate or granular basis. Was there a delay in moving the panel off the line? Was the next panel not ready? Did nail coil reloading take longer? Was there a pause due to rework?
Armed with this information, it becomes easier to rally the production team to focus on the lowest hanging fruit to improve overall line performance.
Hopefully our approach makes sense – we are always open to new ideas on how to make these systems even better. Reach out when you’re ready to discuss using data to improve your productivity.