The Realities
of Implementing a Warehouse Management System
By
Brian Hudock, Partner, Tompkins Associates
Are you one of the companies still running the warehouse
on the host system inventory module or on an outdated locator
system—or even tracking materials on a spreadsheet? There's
no need to raise your hand here in public, but you know who
you are—and you know it is well past time to enter the
21st century and move up to what it requires to remain competitive
in the marketplace. Your customers and supply chain partners
probably know you need to do it, too.
Whether
you choose to upgrade an existing system, add or activate
a new module in your ERP system, or go with a best-in-class,
stand-alone warehouse management system (WMS), you stand to
gain great benefits, and they are generally easy to justify
and to realize if you do the project correctly. The key is
full-time project management and support not only for selecting
a system, but for implementing it and training your people,
too. It is true regardless of whether you will be implementing
a basic locator system, a mid-tier WMS, or a Tier One WMS
with all the bells and whistles to help you get every last
ounce of wasted labor out of your operation.
It can
be in-house management, but the catch is that few companies
have the skill and experience to successfully implement the
technology associated with warehouse management systems. Reading
articles and books, listing to the software providers' sales
pitches and attending seminars does not prepare you for the
reality of implementing a WMS. The only way to avoid the mistakes
that every first-timer makes is to know more than a first-timer
and to be 100 percent dedicated to that project and that project
alone. To help you get there, let's review the most common
mistakes made in WMS implementations.
Mistake
No. 1—Unrealistic Software Expectations
No matter
how simple you think your demands on a WMS must be, there
is no off-the-shelf package that does not require some degree
of customization. If the software cannot be customized to
your operation, then your operation will have to be customized
to work within the parameters of the software.
Solution
Focus
your vendor selection on providers who know the industry and
with system capabilities that match your operational requirements.
This can be done by leveraging outside expertise, and it will
minimize customization. Also, if you are willing to invest
the capital and time to do the preliminary design with the
selected vendor, you can minimize lost time later in the process.
It's important to remember, too, that even after you select
a vendor, the time to hammer out a contract usually drags
out as it passes through legal departments. The bigger your
operation and the more expensive the system, the longer this
takes.
If your
schedule turns out to be unrealistic, adjust it as soon as
possible and spread the news as soon as possible to the entire
project team. The faster everyone can adjust, the lower the
temptation to expedite the system development at a higher
cost or to sacrifice functionality in order to get it done
quicker.
Mistake
No. 2—Trying to Get High-Price Results From a Low-Price
System
The simple
truth is, if you buy low-end software, you get low-end functionality.
Software bargains simply do not exist. Any system will require
some level of modification to properly fit your operations'
requirements, and this adds to the price. Although software
today is more driven by user-configurable system parameters,
those only affect standard operations. Deviations from the
standard set-up still require modification.
Solution
Carefully
identify your operational requirements and prioritize them
both by the financial benefits and the by the qualitative
benefits to customer service and accuracy. Then determine
the level of system you can afford.
Mistake
No. 3—Failure to Track Vendor Progress
It is
a terrible thing to take advantage of a customer's trust,
but the simple truth is that software providers historically
have been able to sell more systems than they can develop
during any reasonable time frame. The old saying is that the
squeaky wheel gets the grease. If you or your consultant are
not monitoring your system developer closely, another system
may be getting the grease.
Solution
No. 3
Assign
a full-time project manager who understands both the technology
and the project goals to monitor the system developer and
to establish milestone tests. This will generally draw delays
and problems to the surface, where they can be dealt with
easily and sooner rather than later.
Mistake
No. 4—Failure to Develop a Contingency Plan
Assuming
that the system will work as designed the day you start it
up and ignoring all the possible negative outcomes is a common
and costly error.
Solution
Make
sure you allow for the system to ramp up. Do not expect it
to be 100 percent efficient on day one. Develop a contingency
plan before start-up to account for not only a total system
failure, but for specific, critical functional areas to go
down. Simple contingency plans include creating back-up electronic
copies of all downloaded files, backing up the system before
start-up and creating a stable recovery start point. New systems
actually tend to hurt productivity in the first month of operation
and then begin the climb to meet promised efficiency over
the next several months. Training and productivity measurement
are a key to making this happen as quickly as possible.
Mistake
No. 5—Overselling the System to the Users
The natural
tendency of the system design team is to promote all the results
that a fully de-bugged system will provide—eventually.
The amount of work required to adapt to new procedures and
to be directed by a computer and the number of irritating
minor bugs that plague most system start-ups are somehow lost
in the sales pitch.
Solution
Everyday
users must be trained in the daily operating environment,
about the changes that will take place, and on the reality
of system start-up problems. Daily users will make or break
a system installation. Spend as much time selling the daily
users on their input and winning their cooperation during
the debugging stage as you do convincing them how the system
will improve their jobs. Let them take ownership of the system,
and they will strive to make it succeed.
Mistake
No. 6—Lack of System Training
Too many
companies believe the software vendors' pitch that anyone
can learn the system in 15 minutes, and they fail to make
training part of the implementation plan.
Solution
Competency-based
training for all system users is critical for system success.
Employees should be given system exposure and training before
system start-up to build both their comfort level and their
functional understanding of the changes that await them. During
final testing and start-up, users must be tested on their
WMS operational knowledge, and those who need it must get
the appropriate re-training. Generally, the best people to
provide this training are other internal users, outside parties,
or vendor training personnel who are not programmers.
Mistake
No. 7—Poor Data
The easiest
way to condemn your WMS to a certain and spectacular failure
is to provide the software vendor with faulty or incomplete
data or to fail to update product characteristic data at all.
Solution
Producing
clean and accurate data is the responsibility of the user,
not the developer. A great WMS will produce lousy results
if poor data was put into the system. Devote the proper attention
to collecting and updating all product characteristic data,
all product pack-size dimensions and the starting inventory
locations of all products.
Mistake
No. 8—Automating a Broken Warehouse
There
is a common misconception that installing a WMS to improve
information flow, entry, and tracking will eliminate all the
inefficiencies within an operation. In reality, improving
only the information flow gives a high probability that the
operation will not even approach the promised improvements—and
it may even lose efficiency. Then whoever championed the WMS
will be updating his or her resume.
Solution
During
the initial definition of the WMS functional requirements,
document what you should be doing rather than what
you are doing now. You must keep in mind the best practices
for your industry and strive to achieve them. To realize the
awesome potential power of change, you must first be cognizant
of what is available. There are two ways to go about it: educate
your organization through seminars, vendor-arranged site visits
and lots of reading; or utilize the expertise of a consultant
to stimulate the process.
Mistake
No. 9—Pointing Fingers
A WMS
start-up is like a rocket launch. Even when it goes right,
there are initial and ongoing critical adjustments before
the payload reaches a stable orbit. And when something goes
wrong, it can go very wrong. What does that have to do with
pointing fingers? Finger-pointing is a normal reaction if
expectations are not met—regardless of whether the expectations
were naïve. What governs is fear that orders will be
lost or delayed, that the company has just flushed a whole
lot of money down the drain on your recommendation, and that
the WMS provider is incompetent.
Solution
During
a WMS start-up, there are lots of adjustment to make in order
to achieve full functionality. They take time, patience, faith
and skill. If it looks like things are going wrong, stay calm.
Work with the software provider and consultants to identify
the true problem, then to find its cause. Once the problem
has been defined, examine the extent of the damage to data
and correct it. (If you made the appropriate contingency plan-Solution
No. 4-this will not be a problem). Do not assign blame; assume
responsibility for success.
Mistake
No. 10—Failure to Measure System Performance
Why spend
the time designing a system and justifying its benefits if
you are not going to confirm that it worked? One of the biggest
failures in WMS implementations is not verifying that the
system is working as designed. There are enormous benefits
if you capture as little as 10 percent of unrealized payback
on the original design-but even that pales compared with what
you can get out of the system by improving the original design
after a WMS has been operational for several months.
Solution
Develop
performance measurement tools and tracking mechanisms to use
for the first six months. Have a manager track them regularly,
report the results and suggest changes to the system.
The
Bottom Line
The best
way to dodge all these problems is to dedicate enough time
to the project and enough resources, whether they are internal,
external or a combination. You will be well-advised to remember,
too, that these are just some of the most common mistakes
made during WMS implementations. Each company encounters snags
unique to its operations. Start by learning from the mistakes
of others and work to obviate those, then keep a careful watch
for your own special pitfalls and potholes-and make sure you
have the best talent you can find to help you avoid them.
You have a lot to gain!
|
