Summary:
Cycle Labs built and ran a 30-45 minute performance test that pinpointed bottlenecks and helped a 3PL validate WMS readiness before Go Live.
Challenge
A 3PL organization providing logistics services such as freight forwarding, intercontinental ocean and air freight, and associated supply chain management was having issues with freight forwarding at a newly deployed warehouse in the northeastern United States, which was about to go live while cross docking international goods.
At this location, international goods are received, staged for customs, processed by customs, optionally packed, staged for shipment, and then shipped. On any given business day, the site was expected to handle a volume of 4,000 cartons.
At that scale, the 3PL was experiencing a variety of inconsistent, impactful issues and couldn’t identify and resolve the root causes, which caused significant delays in the deployment timeline.
Solution
The client engaged the Cycle Labs team for a one-week assessment, using Cycle’s patented performance testing capabilities to help resolve these issues.
Within five business days, Cycle Labs developed and executed a 30-45 minute automated performance test that:
- Dynamically generated Advance Shipment Notices (ASNs) and injected them into the system under test.
- Emulated 20 concurrent terminal device sessions to receive 2,000 cartons into a receiving staging location, then perform a full inventory move to the next system-calculated location.
To produce measurable results, Cycle Labs worked closely with the on-site deployment team to build custom SQL queries that checked whether the data processed by the performance test was in the correct state.
The first run of the test uncovered a variety of quantitative and qualitative issues. The most significant were:
- 3.5-20% of received cartons were sporadically directed to an incorrect location instead of the correct conveyor location.
- Even at the low end, 3.5% of the daily volume of 4,000 cartons is 140 cartons per day needing human intervention to troubleshoot, which amounts to an entire additional staff member.
- 100% of cartons intended to be put away to stock were directed to a conveyor location instead.
- Latency on one specific screen when receiving cartons averaged anywhere from 15 to 60 seconds.
An on-site systems administrator traced these issues to a Java Heap setting. After the setting was adjusted, the performance test was run a second time, and each significant issue had been addressed:
- Received cartons were no longer sporadically directed to incorrect locations.
- Cartons intended for stock were directed to the correct locations.
- Significant latency when receiving cartons was no longer observed.
Cycle Labs also uncovered and reported additional information about remaining issues, but this set of improvements significantly reduced the risk of the warehouse Go Live.
Results
Despite the remaining issues, the 3PL was able to make a strategic decision about what constituted cost-effective risk mitigation and reasonable issue resolution for its deployment.
If you’re concerned about performance issues with your deployment or upgrade, a tactical deployment of Cycle can mean the difference between success and failure.
Download the Full Case Study Here
