Figure 1: Layout and addressing scheme of the fast-pick line
See the photos and accompanying description.
To prepare for expected increases in volume, ABG wants to improve the productivity of its main picking area, the fast-pick area for less-than-carton quantities. This area is configured as an aisle of carton flow rack, supported by a conveyor, and controlled by a pick-to-light system.
Figure 1: Layout and addressing scheme of the fast-pick line
Figure 1 shows the layout and addressing scheme. The fast-pick area is laid out like a backward letter J (with no pick locations on the bottom of the J). The line is operated using the red bay storage/pick locations 11-66. The case flow shelves are 5-feet deep above the picking conveyor and 10-feet deep on the opposite side of the pick aisle. The pick line is typically operated with six zones: bays 11-14, 15-21, 23-32, 33-42, 43-52, and 53-66. On high volume days the aisle is divided into seven, eight, and more zones so that more workers can be devoted to picking. These zones have been pre-determined based on historical data so that on average an equal number of picks is expected from each zone. One picker is assigned to each zone and each picker is evaluated based on two metrics: average picks-per-hour and average errors-per-pick. Pickers are rotated everything few days so that they get experience in each zone.
Management says that the system was designed so that “the first zone drives the line”.
Each carton is considered an order and is presented to the first picker in one of three carton sizes, with a label on the carton that displays the order number. The pick-to-light system will display the order number being picked in the zone and will light up to display required quantities in front of the products to be picked.
After all products from a zone have been picked and the corresponding lights have been extinguished, the picker will push the zone-order-complete button and pass the order to the next zone. The order continues down the pick line in this manner until is complete, when it is pushed onto the early exit conveyor.
An order may not have any picks in some zones and may be completed in any zone. There are typically 5.9 pieces per order, with 94 percent of the pieces picked from the fast-pick area. There is a slow-pick area where low volume product is batch picked and placed in the appropriate carton at the end of zone seven. The slow pick area picks less than 5 percent of the pieces.
The company data is copyrighted and proprietary. You may use it for the purposes of this course only. If you would like to use it for something else, please contact me to discuss.
The data appearing here was from August 2007 and will be updated for the January 2008 projects.
In the data below, each box (shipping container) is considered a separate order.
The project for August 2007 was to recommend whether to convert order-picking in the fast-pick area to bucket brigades. The class argued in support of this and ABG is now converting to bucket brigades. As of this writing, productivity has increased with the first steps toward bucket brigades.
The project for January 2008 is to slot the fast-pick area to account for many goals and restrictions, some conflicting, such as:
In addition, ABG would like suggestions for increasing efficiency of restocking the fast-pick area.
A final question is whether there are efficiencies to be gained by sequencing the boxes through the fast pick area (while ensuring that all boxes using the same mode of transportation are picked together). For example, you might consider picking the smallest orders first. A more complicated sequence is to pick first the orders consisting only of SKUs located toward the end of the aisle. Pushing this idea further, you might consider the aisle as a 2-machine flow shop and sequence the orders as in Johnson's algorithm to minimize makespan.
Read the chapter in the textbook on pieces/eaches and the website on order-picking by bucket brigades.
You will probably want to use the carton slotting tool. Even though it does not do everything the client wishes for, it can give a solution that you can then modify. For example, you can identify all heavy items as a common subgroup so that they will be placed together on shelves reserved for heavy items. Then you can move these shelves in the solution as you wish by simply changing their addresses.
ABG dimensional data is imperfect. You will likely have to go to the warehouse and measure several hundred SKUs with missing or suspicious dimensions.