Anyone who's bought a cell phone lately understands the idea of convergence. Where we once made calls on a phone, listened to tunes on a portable disc player, snapped pictures with a camera, and surfed the Web on a laptop, we can now do it all on one device.

Something similar is beginning to happen in the warehouse and DC.

Bar codes no longer rule the day. End users now have a variety of technologies to choose from to execute the processes managed by a warehouse management system (WMS), including voice, RFID and camera-based imaging.

For the most part, different devices and software interfaces are required to manage these technologies. But ADC technologies are beginning to converge onto one multi-modal device that can handle them all.

• Want to talk to the WMS system? No problem.
• Need to scan a bar code? Sure thing.
• Need to read an RFID tag? You can do that, too, without carrying a separate device.

Just as importantly, the next generation of WMS is being written to include core functionality to manage those technologies without middleware.

“The idea behind a multi-modal warehouse is one of making data collection and WMS technology serve the operator,” says Dick Sorenson, director of RFID products for LXE. “You want the option to choose the technology that minimizes data collection time, reduces errors and naturally improves that particular process.”

While this is the vision for the future, it's not quite there yet in most cases. “One of my industry contacts pointed out that WMS were architected years ago around bar code scanning because that's all there was,” says Steve Banker, service director of supply chain management for ARC Advisory Group. “If you were building a system today, you would build it to be multi-modal. WMS vendors recognize that and are moving in that direction, but it's not an easy process.”

The adoption of voice is the first step most facilities are taking toward using multiple data collection technologies. According to a recent survey of Modern readers by Banker and ARC, the adoption is happening at a fast pace. “Five years ago, if a warehouse was choosing between voice and pick-to-light technology for each picking, the choice was usually pick-to-light,” says Banker. “Today, the choice is more often than not speech recognition.”

Most often, voice is being adopted as a stand-alone technology that replaces bar code scanning, especially in order picking applications. “The benefit is increased productivity and accuracy,” says Scott Yetter, CEO and president of Voxware. Yetter says productivity improvements of 10 to 25% are not uncommon, as are error reductions of 50%, especially for users moving from paper-based picking to voice.

But voice is also the first data collection technology being combined with bar code scanning to create new, multi-modal processes. “We have customers picking products to a pallet,” says Mike Glatz, director of business development for Vocollect. “But they also have to capture a lengthy lot number. Voice will direct them to a location and tell them what to pick. Then they will scan the lot number because that's faster than speaking a nine digit lot number.”

Other users rely on a combination of voice and scanning to verify order accuracy. “We have a retail customer doing voice-directed picking that still uses bar code scanning for order verification and quality control,” says Kevin Prouty, senior director of manufacturing solutions for Motorola. “What they have found is that there is still that 1% of the time that for whatever reason, the voice command didn't work.”

However the technologies are combined, going multi-modal changes the way you approach operations. “You can now look at your warehouse and determine which method or combination of methods makes the most sense for a process,” says Banker.

Still, there are hurdles. While the device might be multi-modal, there are typically separate software solutions for the bar code scanner and the voice technology. An operator can't arbitrarily choose between speech and scanning in most WMS, at least not without getting out of one system and into the other. And that takes time, which could nullify the productivity benefits of a multi-modal approach.

That's because there have traditionally been different approaches to WMS and voice technology, according to Banker.

Approach 1

Simply bolt a voice solution onto the WMS, rather than integrate the solution. This might be the case for end users with legacy or best-of-breed systems implemented before voice became common. In that instance, the voice system receives order instructions from the WMS, then updates the WMS in batch mode after the tasks are completed.

The downside: “With a bolt-on solution, you lose some of the efficiency you would otherwise get from a system-directed RF system that is integrated with the WMS,” says Tom Kozenski, vice president of product strategy for RedPrairie. “For instance, if the WMS isn't directing the operator, it won't know in real time that a slot is empty and needs replenishment.”

Approach 2

Leading Tier I WMS providers, like RedPrairie and others, partnered with a supplier and integrated their voice software system with the WMS. That overcomes the integration issues and results in easier configuration. “If you've integrated a speech engine and have a workflow-based WMS, you can define whether a specific task is going to interface with a bar code scanning device or a voice device,” says Chad Collins, vice president of global strategy for HighJump Software.

Approach 3

Just coming to the market are WMS suppliers that have built their own voice engines and incorporated those into the architecture of their WMS solution. “It's no longer a bolt-on or another piece of third-party software integrated into their solution,” says Banker.

A true multi-modal WMS. “You now have software for mobile workers that runs on a mobile device,” says Mike Markham, vice president of sales for Cadre Technologies. “You can capture data or communicate with this device in as many ways as you want without deciding ahead of time in a workflow configuration what input method you want to use. I can key it in; I can speak it; I can scan it—all without making any changes.”

Markham makes a distinction between a multi-modal WMS and a multi-modal device that can scan a label, read an RFID tag or communicate with speech.

“With a multi-modal WMS, the application allows you to make any kind of input you want without having to manually switch from the voice application to the scanning application,” he says. “Because the solution is part of the core WMS functionality, picking instructions, for instance, are delivered via speech and the screen at the same time.”

The first truly multi-modal applications using multi-modal devices, like a PDA with voice and bar code scanning functionality, are just beginning to be developed, adds Donal Mac Daid, product marketing manager for Aldata, one of the other suppliers to create a multi-modal WMS. “We have a customer in Europe that sends a pick list to the operators' screens,” says Mac Daid. “But an operator can reorganize the pick list using voice technology as she is picking.”

For one, you have just one terminal to purchase and maintain. Beyond that, it creates opportunities to more fully integrate data collection throughout a facility.

“It's opening up the possibility of developing new processes and rethinking the traditional WMS environment,” says Mac Daid.