GHQ: Out of the Box
By David Litwak
Through special arrangement, what follows is an excerpt of a current story from Grocery Headquarters magazine, presented here for discussion.
Warehouse management systems (WMS) have become almost universally popular over the past decade. At first, manufacturer- and retailer-owned warehouses, as well as distribution centers, often either developed their own systems and software in-house or hired a consulting software company to develop something specific to their needs. Unfortunately, these proprietary systems were usually not robust enough to handle the increasing complexity of interconnectivity with the systems of other companies along the supply chain.
That caused the momentum to quickly shift to software developed by third-party suppliers that offered a degree of standardization among supply chain partners. Subsequent generations of WMS software have stressed their ability to be customized to specific needs and operations through a wide assortment of modules and programming written specifically for each user. Makers of many systems have also stressed the ability of their programs to handle many different phases of transportation management beyond the warehouse.
Does such a high degree of customization mean the industry has begun to return to the days when each company with a warehouse was basically running its own proprietary WMS software? And has this customization led the industry to a point where implementation has become a lot more complex and the programs may have lost some connectivity with the systems of different supply chain partners? To some industry executives, both those developing WMS software and those running it, the answers to these questions are yes and yes.
There is a group of managers who feel that the best way to corral the functions of a large, highly complex warehouse operation is to implement a comprehensive WMS program that covers as much of the operations in a standardized manner as possible. The less customization the software needs, the easier the implementation.
According to Ann Price, president of Motek, overly customized and complex WMS programs have had a negative impact on the efficiency of warehouse operations. Some of the spectacular failures to which complex WMS systems have contributed, she said, included errors in Nike’s inventory management software project that cost the company $100 million in one quarter and the failure of J. Sainsbury’s $700 million automation project, which led to the replacement of most of the supply chain team.
While the idea of an out-of-the-box solution to control the increasingly more complex operations of a distribution center or warehouse is appealing to almost every manager, many still feel that their unique set of operational needs cannot be adequately met by a primarily standardized solution.
Customized programs have the upper hand for warehouse operations that have variables or specific products and situations that are beyond the scope of the standard programs. For the truly unique operation, a WMS with customized components is a better approach. The one thing that most executives do agree on is avoiding a system that is completely customized, which smacks of the legacy approach of the past.
The great unknown when picking a WMS is the future. The frequency with which the logistics industry is changing, in terms of technology, function, structure and composition, seems to be increasing exponentially, which means that its future direction is hard to predict. Equally difficult is guessing what will be expected from a WMS. As the operations of the company warehouse changes and widens, will the standardized WMS or a more customized one be more advantageous? New software releases in the future will hopefully be able to cope with the warehouse changes ahead.
Discussion Questions: Does the high degree of customization in warehouse management systems mean the industry has begun to return to the days when each company with a warehouse was basically running its own proprietary software? Has this customization led the industry to a point where implementation has become a lot more complex and the programs may have lost some connectivity with the systems of different supply chain partners?
When installing any software package (warehouse management, point of sale, merchandising, financial, etc.) it pays to (1) find the product that fits the best and (2) refrain from customization until you’ve used it so long that you know it well. When a man buys a suit, it pays to look for something that only needs minor alterations (cuff length, not the shoulders or the collar width.)
Customization is needed inside an operation because companies have developed solutions and efficiencies that are often unique to what happens inside their own facilities. However, standardization can benefit any supply chain entity that chooses to connect in such a way outside the box. The benefits that flow into the DC along with synchronized and standardized communications should be the next level of great rewards. One example: passing product expiration dates into the distributor WMS could result in more accurate inventory management and purchase decisions that reduce the incidence of expiration.
The Global Commerce Initiative with the GS1 and GS2 groups are the standards-setting bodies for data standardization to make synchronization and data exchange more efficient. The marketplace problem is that suppliers created software that provides many of the needed functions; some of the software has much opportunity for customization. However, in the rush to get products on the market, suppliers create what they think warehouses will need. When deciding upon warehouse software, managers need to keep their needs and goals foremost in their minds when evaluating and choosing software. Usually they pick the software that comes close. If that software uses global standards and is customizable, the choice should be workable. The two critical issues are whether they software is compatible with global standards because that will improve the possibilities of being able to synchronize with partners and whether there is some ability to customize the software so that managers can ensure that the software will meet the goals of the company.
It’s hard to know where to begin with this one….
Yes, there is still a lot of customization in most “high end” warehouse management system implementations. However, this customization is still a far cry from the hugely custom systems that characterized the implementations until fairly recently.
The customer/user is always chasing more advanced features. So, while what was custom written in the past becomes a standard “out of the box” feature today, the user now looks for more advanced capabilities that aren’t out of the box.
Warehouse management systems, like ERP, are also to some extent a victim of their own growing sophistication. That just means, as the programs becoming more and more functional and complex out of the box, relatively minor changes require more and more custom hours to implement, because each function touches so many things.
So, netting it all out:
1. If you have much automation (conveyors, etc.) in your warehouse, as most retailers do, you will have a decent amount of customization to deal with that, regardless.
2. As warehouses and their drive for increased efficiency move relentless on, users will push for capabilities that just still aren’t yet part of the package.
3. As packages grow in complexity and scope of add-on modules, relatively small changes become costly because of all the touch points.
4. The total functionality being delivered today is well beyond what was received 5 years ago. No question about it.
There is also a trend, somewhat alluded to in the story, back away from the attitude we had for awhile, which was companies were thinking it was best to modify their processes to avoid customization. That attitude has been swinging back the other way for a bit, though not totally.
The key, as I have personally experienced many times, is to know what unique processes are truly adding value, and hence worth customizing, and what are just “the way we’ve done it,” and you are paying to pave over cowpaths.
There are four major perspectives that come into play when considering WMS applications: physical layout of the warehouse, level of automation, product mix and customer mix. I often tell prospective users that the most important factor in their distribution decisions is the size of their stores’ backrooms. Ultimately (hopefully), all the products you have bought will end up in the stores.
As the variety of options on each perspective continue to increase it becomes more difficult to locate a WMS that meets all the aspects of a particular situation. Do you need to support multiple buildings, automated pick machines, hazardous material storage, or allocate inventory to various customers? Those are just one example of the special requirements that might exist in a particular implementation.
Potentially, this is where a Service Oriented Architecture (SOA) solution should provide the necessary foundation. This means locating a base application that contains the “goes-into’s” where supporting applications can be plugged into the overall process. Some of this is covered by the ARTS models which provide a basic understanding for a WMS. I don’t know of any vendor who has really designed a WMS with published SOA interfaces that would allow additional service providers to plug into their process, but I suspect they exist.
The fundamental question here is whether we will ever get to the point where any application can be generally accepted without any customization. I think the answer is “no.” But I also believe that Pareto’s Rule applies and that a great deal of the benefit can be achieved by locating solutions that cover 80% of the need while supporting simple customization techniques for achieving the additional 15%. That last 5% will come from the most flexible piece of your operation; the people who run the applications will adapt.
To some extent this brings up the old conflict between business and technology decision makers. The best choice for a WMS may have nothing to do with the current features which have been so highly touted to the potential user, but with the internal configuration of the application and its ability to accommodate new features in the future. The feature rich solution and the technically flexible solution are not necessarily the same.