Many manufacturers and distributors know a paperless warehouse management system (WMS) could cut costs by increasing productivity and accuracy, which also improves customer service. But a good WMS (involving bar coding and RF scanning and, soon, radio-frequency identification reading) is more complex and expensive than most people have been led to believe. To avoid installation problems, follow these upfrontplanning steps to select a cost-effective WMS.
Involve top management. Principals and executives of many manufacturing and distribution companies believe anything involving computer systems or the warehouse is best left to others. Not so. Most successful military campaigns resulted from the leadership of a general, even if he remained well behind the front line. If the effort to select and install a WMS is to be successful, top management must provide the leadership, even if that means learning something about computers and WMS.
Organize a team. There is a natural tendency at most companies to limit the WMS search team to warehouse and MIS personnel. They should be part of a team, but the team should include people from customer service and purchasing, because the system affects them.
Estimate growth and change. Since a WMS is a long-term investment, it's important it be able to handle future and current warehouse needs. That means estimating such future operations as orders and line items picked per day, average and peak and daily putaway volume.
Tighten procedures and controls. Automating a weak manual system results in an expensive weak automated system. Make sure needed procedures, like receiving and putaway, and controlsfor information and product flow are definedand being followed.
Evaluate the enterprise or management resource planning (ERP/MRP) system and its use. Every WMS depends to some degree on the main business system, the system used for sales order entry, inventory tracking, and purchasing. The main business system must have functions and data that are "expected" by a WMS; for example,expected arrival date of a purchase order. The data in the business system must be up-to-date and accurate, and users must know how to use it effectively. This determination is needed even if the provider of the ERP/MRP system is selling a WMS. If many problems are discovered during the evaluation, especially problems with the business system (which will require software changes or additions), slow the search for a WMS.
Estimate costs. Although WMS software and training can be expensive, RF and bar code equipment is not cheap and shouldn't be taken for granted. Equipment ("access points," hand-held guns) should be generously estimated and used to obtain cost quotes.
Estimate savings and benefits. Be conservative when estimating personnel reductions, personnel avoidance, and other out-of-pocket savings, such as less-damaged product resulting from fewer mispicks and returns. Don't ignore the impact of non-cash benefits, such as happier customers resulting from a faster way to handle customer returns.
Selection and installation Define detailed long-term system needs. Before going shopping, create a detailed list of specific current and future WMS requirements, and equipment needed. Without such a list it's impossible to judge whether a particular WMS contains specific needed functions and impossible to understand vendor proposals and compare different systems. Do this even if the only WMS under consideration is the one provided by the vendor of the MRP/ERP system.
Solicit written bids. Ask WMS vendors to compare their software against the list of requirements, feature by feature, even if the only vendor involved is the one that provided the business system. A written detailed comparison is needed because there is never enough time in a demo to explore all the functions on a comprehensive requirements list.
Investigate each system. Just as some people select a WMS solely by observing demos not by soliciting written quotations, others make the mistake of basing their selection solely on an examination of vendor quotations (skipping demos and other important steps). Without a demo, it's impossible to judge whether a system has an acceptable look and feel and whether the various features are practical to use. And, even if a demo takes two or more days, use that opportunity to have users try out various features and functions. After each demo, the team should decide if the system is worth pursuing further; if so, call a few references of the vendor. After the telephone reference checks are completed, the team should decide whether that system is so attractive that a visit to one or two references is worthwhile.
Select the most cost-effective WMS. Two kinds of facts are needed to make an informed decision: financial and non-financial. The biggest financial fact to consider is long-term cost of ownership, the initial and ongoing costs (including costs of having the vendor modify the WMS to interface with the business system). Among the many non-financial facts is the degree to which the unmodified WMS software fits the company's needs. Many of the non-financial facts can be obtained only by studying a written response to the requirements list. The big decision must be made taking both financial and non-financial facts into account.
Add specific performance guarantees to the contract. No matter how comfortable the team feels about the selected vendor and system, installing a WMS is so complex that many things can go wrong—before, during and years after installation, sometimes with a costly impact. To protect against problems, a contract should list everything the vendor promised and everything the team is expecting; in the form of an addendum, if the contract form cannot be modified. Some of the many issues that should be addressed in a contract include: exactly what is being provided, and at what cost; what happens when there is an equipment failure; what happens if a "bug" is found in the WMS software. If the vendor refuses to even negotiate the changes, find another vendor. If the vendor objects to certain terms in the addendum, the team leader should ask the vendor to explain why and suggest an alternative. Be prepared to compromise.
Integrate and test software before installation. As much as possible, before taking delivery, have the WMS vendor change the WMS software so it integrates with the business system. Before delivery, test the modified WMS software—assuming that the business software has been changed to accommodate the WMS (as described earlier). Use real data for the test, which should be conducted by the people who will be using the new WMS.
Don't skimp on user training. Because the cost of a modern WMS often produces "sticker shock," most companies ask for a discount on hardware and software, and cut back as much as possible on related services such as training. Big mistake. A WMS is so complex that the only way to learn more is to spend a lot of time in formal classes and on-site training sessions, all of which cost more than most people expect.
Dick Friedman, president, General Business Consultants. Contact him at 847-256-3260; http://www.genbuscon.com/
