Oct. 2 2006 03:16 PM

When working on the implementation of a Warehouse Manage-ment System (WMS), nothing beats experience. The only negative part of that concept is the pain that one has to go through in order to get that experience. So that you might avoid some of the grief, here are a few lessons that I have recently learned from some particularly painful implementations.


Lesson 1: You Know What Happens When You Assume the Details of Package Integration

Integration between any two packages can be a tricky process. Specifically between a WMS and a host ERP, it can be downright treacherous. Rule #1: Never assume packaged interface modules are everything the salesperson tells you. You are often well into the implementation when you finally roll back the covers and see what interfaces need to be modified or created. At this point, it is crucial that all involved parties focus on the effort required, avoid finger pointing and make effective communications a priority.


On a recent project, integration work was being done by a software provider for its host system. Even though it was a critical path item, our progress on package integration was incredibly slow in the beginning in large part because of communication problems. The end users IT department insisted that all communication to its host system software provider go through it. This added avoidable delays to the process. Another issue was the fact that 3,000 miles and three time zones separated the parties working on the integration. The package integration effort made huge strides when it was decided that both parties (project team and host software provider) be in the same place working on the interfaces. Even though there were unexpected travel costs involved, a host software resource was brought to the west coast facility, and in a couple of weeks, more progress had been made than was accomplished in months.


Lesson 2: Dont Underestimate the Value of Field Acceptance Testing

I was recently reminded about the value of extensive field acceptance testing (affectionately known by consultants as FAT). User acceptance testing (UAT) had been progressing with some mixed results. For example, test scripts needed to be re-written, data setup proved problematic, and our designated super users did not seem to be picking up a lot of knowledge. So the decision was made that only two rounds of UAT be conducted (instead of the originally planned three) and more emphasis be placed on starting FAT right away. This was very successful and flushed out a lot of things that we would not have found with more UAT.


The FAT provided enough stress to the clients network to highlight the fact that the additional load would start causing network problems. Earlier testing had already revealed a faulty core switch in the network. Also, the client wanted some assurance that reasonable throughput could be expected after go-live. So after the first few rounds of FAT, an additional round of volume testing (in reality another FAT) was conducted. In total, we conducted FAT five times, including the volume testing. This testing not only validated expected throughput but also provided a lot of practice for operators and served as a great training vehicle. To highlight the value of learning through hands-on testing, I remember an operator on another project who had been through two rounds of RF training in a classroom setting. When out on the floor to help with system testing, RF unit in hand, he looked up with a blank expression and said, What do I do?


Lesson 3: Use Experienced Project Management Resources to Minimize Risk

All successful projects are based on realistic expectations. Even though acquiring experienced project managers may be costly, they are worth their weight in gold in order to minimize risk to your business. An experienced project manager will realize that you really cannot apply a cookie cutter approach to each project. Even though the WMS may be the same, and the integration resources are the same, you must understand that each implementation has its own unique set of challenges. Also, the experience will allow your team to avoid the all-too-common landmines that exist as you move towards go-live.


Lesson 4: Look for Opportunities to Slice and Dice the Implementation

Finally, whenever possible, look for opportunities to slice and dice the implementation to reduce risk. A recent project was originally planned to go live with the whole facility all at once a risky venture under the best of circumstances. In the midst of the project, a decision was made to focus on the e-commerce fulfillment facility first then attack the wholesale facility later. Now that the e-commerce facility has been launched successfully, we are planning on bringing up a portion of the wholesale business before the remaining portion once again, splitting the implementation to minimize risk to the business.


Hopefully, these lessons will be of assistance to those of you who are setting out to implement a WMS. Learn from the mistakes of others. Ive certainly learned from my own. As my dad used to say, if a dog bites you once, its the dogs fault; if a dog bites you twice, its your fault.


Todd Lawson is an experienced project manager and systems implementer, with a specialization in Warehouse Management Systems selection and implementation. He has a Bachelor of Science in Industrial Engineering from the University of Tennessee, Knoxville and is a member of WERC. He can be reached at TLawson@Q4Logistics.com.