The Journal of Business Logistics (JBL), published by the Council of Supply Chain Management Professionals (CSCMP), is recognized as one of the world's leading academic supply chain journals. But sometimes it may be hard for practitioners to see how the research presented in its pages applies to what they do on a day-to-day basis. To help bridge that gap, CSCMP's Supply Chain Quarterly challenges the authors of selected JBL articles to explain the real-world implications of their academic research.
THE ARTICLE
"Just-in-time retail distribution: A systems perspective on cross-docking," by Paul Buijs of the University of Groningen, Hans W. Danhof of the Dutch retailer Blokker, and J. (Hans) C. Wortmann of the University of Groningen. Published in the September 2016 issue of the Journal of Business Logistics.
THE UPSHOT
Cross-docking—the process of moving goods through the distribution network without placing them in stored inventory at a distribution center—typically involves moving products from an inbound trailer directly to an outbound trailer or temporarily storing them on the floor before shipping them out. Cross-docking improves the processing speed of a distribution network while also reducing the amount of inventory it needs to hold.
Many companies, however, struggle to effectively implement cross-docking strategies. One of the main reasons is that most have implemented cross-docking without changing their organizational structure or metrics. Most supply chain literature agrees that the best approach is a holistic one, where cross-docking operations are not only synchronized with inbound and outbound logistics processes but also are managed by the same people with the same or similar performance metrics.
But this does not often happen. Managers who oversee cross-docking operations typically are not also involved in external logistics processes, and the metrics they use focus only on internal efficiencies, such as the distance traveled by material handling equipment in the distribution center.
In this article, the authors sought to prove the extent to which a holistic approach to cross-docking provides a significant advantage over a more localized approach focused only on the distribution center. To accomplish this, they worked with a major retailer in the Netherlands to identify cross-docking improvement opportunities. The two possibilities they studied were 1) whether to change the dock-door assignment policy; and 2) whether the retailer should cluster and sort loads bound for the same store at the cross-dock itself or at a facility farther upstream in the distribution network. They also used simulation software to determine which would create a bigger impact: focusing on local optimization, or focusing on networkwide optimization. As part of that process, the authors discovered that the retailer's current metrics were not fully communicating the benefits of a holistic approach. To address that shortcoming, they developed new metrics, borrowing from concepts used in lean manufacturing.
Dr. Paul Buijs, the lead author, spoke with Supply Chain Quarterly about what these findings could mean for companies that are currently using cross-docking or are thinking of implementing the technique.
What issues were you seeking to explore through this research?
We saw that most of the benefits of cross-docking are lower inventory levels. But the lower inventory levels also form the main challenge of cross-docking. Due to the low inventory levels, a much tighter coupling arises between the logistics inside the distribution center and the inbound and outbound logistics networks. We also saw that both the academic research as well as practitioners' own strategies were geared mostly toward optimizing the operations that take place locally—in other words, how to improve cross-docking operations at the distribution center itself. But due to the tight coupling between the local cross-docking operations and the network logistics, there is only so much [benefit] you can [achieve] when you have this localized focus.
Our main message is that future efforts should be geared toward addressing cross-docking from a holistic perspective where we consider local and network considerations together. But first we wanted to empirically verify this claim. Then we also set out to provide a detailed example of what firms could actually do with this holistic approach, what it generally implies for managers, and how they can adopt it.
Your paper presents a case study of a large retailer in the Netherlands. Why did you choose to focus on this particular retailer?
First of all, this retailer is one of the largest international grocery retailers, and it's considered a leader in how it's organizing its distribution. Cross-docking forms a central part of the company's distribution strategy. And at the time we started this research, the retailer was planning a major change in its distribution network. Although that change did not actually relate to cross-docking, it offered a very good opportunity to propose and test some distribution network changes that we thought could improve cross-docking from a systemic or networkwide perspective.
On top of that, one of the warehouse managers at the retailer was willing to cooperate with us in proposing and testing a change for a local cross-dock improvement effort. This provided a rather unique opportunity to gather empirical data to support the call for a more holistic approach to cross-docking.
What makes the performance metrics you propose in the paper different from the traditional metrics used by cross-docking distribution centers?
Metrics was one of the key things that drove this research. We believed that because cross-docking is all about reducing inventory and improving flow, it has a close analogy with lean manufacturing. It therefore also makes sense to make this link with performance metrics.
What emerged during our research is that there was a lack of performance metrics that would trigger management to look at cross-docking more holistically. Cross-docking operations are managed according to traditional warehousing principles, where one manager would be responsible for operations inside the distribution center and another one for the transportation or other logistics at the network level. On top of that, each of those managers would have his or her own set of metrics geared toward either localized performance or networkwide performance.
With these existing metrics, it was very hard for us to convey the need for the changes that we were proposing to the retailer. So we added some performance metrics inspired by "lean" and just-in-time manufacturing that focused on the flow of the loads and work-in-progress throughout the distribution network. An example would be that we kept track of the number of in-process loads that were on-site at the cross-dock, which translates into the work-in-progress metric from lean manufacturing. We also tracked the life span of loads throughout the cross-docking systems [how long it takes a load to go through the distribution network as a whole], which gives an indication more or less of the flow.
We also used more traditional metrics because we felt we could relate more easily to the managers using their own metrics. An example of a traditional metric that we incorporated was the travel distance covered by the material handlers inside the cross-dock. The less time material handlers have to travel, the more efficient the cross-dock operations are considered to be by managers.
How can companies use this information to improve their own cross-docking operations?
Our study shows that while local improvement efforts for cross-docking can be very effective at making the operations inside the distribution center more efficient, the impact of these improvements from a systemwide performance perspective can actually be quite limited.
On the other end, our paper shows that even minor changes in the network design could result in considerable systemwide performance improvements. The network design change studied in our paper involved changing the location at which loads are clustered and sorted for store delivery; that is, from the distribution center to one facility farther upstream in the distribution network. This is just one example of just one kind of networkwide change that could seriously benefit cross-docking performance.
The paper also shows that changes at the network level affect another type of performance metric—metrics that not many firms currently use in cross-docking. Without such metrics, many opportunities to improve operations may go unnoticed. We provide just a few examples of metrics that would reveal these opportunities, such as the number of load carriers [wheeled equipment for moving cases] on-site or how long it takes for the load carriers to move through the distribution network. The data for these measures are typically already available in a firm's existing warehouse management system.
What is the key takeaway from your research for practitioners?
First and foremost, we empirically verified, and therefore stressed, the importance of taking a holistic approach to cross-docking. But in order to take this approach, cross-dock management needs to be organized differently. Firms could consider changing the way management responsibilities around cross-docking are organized. They could also consider adopting new performance metrics that better reflect the holistic approach to cross-docking.
When you have functional silos, with managers who are responsible for internal operations and managers who are responsible for network operations, and they each have their own metrics, it's quite hard to make cross-docking happen efficiently. So a key takeaway is that you need to organize cross-docking differently to see the opportunities and then to seize them.
TO READ THE FULL ARTICLE ...
As a member benefit, CSCMP members can access articles in the Journal of Business Logistics at no charge. To request access to this and other JBL articles, send a request via e-mail to cscmppublications@cscmp.org.
