Company: HK Systems
Location: Milwaukee, WI
Engineering services: Computer/software, mechanical subsystems, and electrical control upgrades, including controllers, motors and drives, operator interfaces, positioning systems, communications and power systems for materials handling automation; System simulation, animation, and emulation
Project: Total equipment redesign and retrofit of Sara Lee Casualwear's Tultex distribution center in Martinsville, VA, including the upgrade of the drives, controls, and optical positioning system of a ten-year-old automated storage and retrieval system (AS/RS). This 11-aisle system consists of a shuttle-mounted pallet extractor that travels horizontally and vertically in a 483-ft-long aisle between 100 bays of 78.8-ft-high, double-deep pallet racks.
Engineering challenge: Boost the capacity of the AS/RS from 5,676 to 6,600 pallet stores/retrieves per 15-hour day. In the existing configuration, a single ac drive controlled both the motor for the horizontal axis and motor for the vertical axis, which operated at either high or low speed. This design resulted in a compromise between the acceleration and deceleration rates of both axes-both were less than could otherwise be achieved. Also, the existing horizontal and vertical positioning encoder system and code plate resulted in excessive creep time to locate the correct pallet position.
Proposed solution: In order to meet the throughput needs, the initial belief was that the crane speed would need to be increased significantly and that servo motors would be absolutely necessary to meet these requirements. We proposed two alternatives: Alternative A involved drive and positioning improvements, specifically new, dedicated variable frequency drives for all three axes of motion (shuttle, vertical, and horizontal), as well as a new, laser-based absolute positioning system. Alternative B consisted of the same improvements to the horizontal and vertical motions, and also included ac servo motors at an additional cost of $200,000.
Analysis: People often erroneously believe that increased speed automatically equals increased throughput in an automated materials handling application. To better understand the activity cycle of the existing equipment, HK's simulation team visited the facility and measured the specific acceleration and deceleration rates of the crane, as well as the pause and creep time associated with locating a specific pallet position. The dual-cycle scenario assumed that the crane would perform the following moves:
Pick up two pallets at the input station
Deposit one load in a deep position in the front half of the rack
Deposit one of the loads in a near position in the back half of the rack
Pick up one load from a deep position
Pick up one load from a near position
Deposit two loads at the output station
Then we simulated this same activity cycle for each of the equipment upgrade alternatives outlined above. The results are summarized in the accompanying table.
Results: Our findings showed that we could in fact achieve the required increased throughputs through improvements in the drives and controls technology and position system alone, without any increase in the crane speed. In fact, given typical travel distances, the crane will not even reach top speeds in many cases. Furthermore, although using more expensive motors reduced the shuttle cycle somewhat, the incremental improvement did not justify the added cost. In this particular application, the ability to increase the acceleration and deceleration rates and reduce pause/creep times was more than enough to meet increased throughput demands.
|Performance parameter||Existing||With drive and positioning improvements|
|Horizontal speed (max)||350.0 fpm||350.0 fpm|
|Horizontal acel/dece||l0.8 fpss||1.0 fpss|
|Horizontal pause/creep||8.0 sec||4.0 sec|
|Vertical speed (max)||65.0 fpm||65.0 fpm|
|Vertical acel/decel||1.6 fpss||1.6 fpss|
|Vertical pause/creep||3.0 sec||2.0 sec|
|Communication time||1.0 sec||1.0 sec|
|Single deep shuttle cycle||16.1 sec||11.6 sec|
|Double deep shuttle cycle||24.5 sec||18.3 sec|