Cloud WMS: A Platform for Reliability, Scalability, and Performance

An Interview with Diego Pantoja-Navajas, Vice President of WMS Cloud Development

Oracle WMS Cloud

ARC Advisory Group, in partnership with DC Velocity magazine, conducted survey-based research on the changing practices, priorities, and expectations of warehouse executives. This recently completed research unveiled several interesting insights. Last week, Diego Pantoja-Navajas, the Vice President of Oracle WMS Cloud Development (former founder & CEO of LogFire), and I exchanged our perspectives on some of the more pertinent findings.

Clint: As an executive responsible for the development of Oracle WMS Cloud, I suspect you will be happy to hear that our survey respondents chose WMS as the one warehouse technology investment that is the highest priority/importance to their organizations over the next three years. I see this as an indication of WMS being a mission-critical application that serves as the backbone of warehouse operations. What are your thoughts on this finding?

Diego: Yes, WMS is certainly a mission-critical application for today’s warehouse operations. While some other technologies can provide step improvements in productivity, WMS is necessary for productivity, accuracy, and as a system of record. Cloud applications running on leading enterprise cloud infrastructure offer extremely high availability. Oracle WMS Cloud, for example, is a zero-downtime application – both for patching and updates. We consider it critical that we perform as well or better than an on-premise application.

Clint: OK. Let’s talk more about WMS application performance. Are there cases where you believe cloud infrastructure provides a performance advantage over on-premise WMS applications?

Diego: There are many situations where a cloud WMS can provide superior performance. Rapidly shifting fulfillment requirements, such as those caused by the recent coronavirus pandemic, comes immediately to mind. Our customers that provide essential products, like supermarkets and pharmaceutical companies, recently had their volumes increase to Cyber Monday levels. The cloud architecture and algorithms allow us to dynamically scale cloud computing capacity to handle customers’ increasing volumes.

Clint: Do you believe on-premise WMS would not provide adequate computing capacity to handle these volume spikes?

Diego: On-premise applications can also perform at these levels, but it would be cost-prohibitive. A true cloud solution offers notable cost per unit advantages.

Clint: Speaking of high-volume fulfillment operations, the vast majority of respondents to ARC’s research currently use intelligent system directed batching and waving to fulfill orders. And 76 percent of those respondents stated that their use of batch and wave processes has increased over the last three years. I believe these results show that warehouse processes are getting more mature while operations are also becoming more complex. What are your thoughts?

Diego: Warehouse operations are becoming more complex and so is wave processing. It is important to enable various wave processing methods that support very high order volume processing both for mobile hand-held based operations and highly automated end-to-end distribution systems. Also, complex operations with high throughput typically make full use of advanced capabilities such as system directed tasking, interleaving, and REST Web APIs for automating receiving, routing, pick, pack processes. Finally, warehousing complexities also differ across industry verticals. So high application configurability is important to support industry-specific requirements.

One of the things we have in our DNA – from our LogFire days – is the fact that we constantly run performance tests to make sure the system is running at its best at all times. In recent testing we found that the speed for order line processing increased 20 times the order quantity while only barely doubling the time taken.

Clint: That is impressive, and counter-intuitive that more calculation is taking less time. Why do you believe that is the case?

Diego: We process records in bulk which improves the average speed. We’ve proven consistently that time performance gets better as we add more and more orders to each scenario. There are two reasons for this. The first is the power of the infrastructure. The second is the fact that running on that infrastructure means we can tune our execution systems for maximum efficiency.

Clint: Finally, our research indicates that the adoption of warehouse automation will continue to increase, and the value proposition of automation will continue to increase as well. In fact, over 90 percent of our respondents stated they expect the value proposition of warehouse automation to increase (relative to the manual warehouse alternative) over the next three years. What are your thoughts on this topic?

Diego: There is an ongoing problem with finding qualified labor in developed countries, and now with Covid-19 and the potential reshoring of manufacturing, that problem has been exacerbated. So the answer is automation — especially considering that beyond certain throughputs automation is the only way to scale.

To help fulfillment operations meet this demand, WMS providers need to build out capabilities, such as operational REST Web APIs that allow integration with MHEs and/or MHAs in a vendor agnostic way to help orchestrate the functioning of all these pieces of equipment within a warehouse. Some vendors like to call it WES. I just like to keep it simple by calling it operational REST APIs for automation – keeping all those features within the WMS, no need for another layer of software.

Robotics and automation services, meanwhile, are also advancing integration capabilities. Our plan is not to compete with these automation providers, but to create a sophisticated ecosystem of automation vendors to serve the market.

Clint: Thank you for your insightful perspective. We look forward hearing more about Oracle WMS Cloud in the future.

Leave a Reply

Your email address will not be published. Required fields are marked *