Voice Picking

I thought I would discuss the picking process further given the number of comments and questions generated from my last blog entry of a similar topic. This time I would like to discuss some challenges and solutions associated with voice picking in the DSD process flow.

Voice picking basically follows some similar steps, the picker receives instructions like “Go there (insert locator ID, whether that be a location code, or sequence #) and pick that (insert product indicator and quantity). ”After completing this task with a product scan and quantity validation, the picker is then given the next set of similar instructions (the details and variations of how to complete this task are for another entry). Typically order header-type instructions will also be provided at beginning or end of these steps. But the problem that I see for many DSD customers is that they have limited visibility as to what product is going to come next. Forget about a user friendly way to see an overview of what is going to end up on this particular pallet. Without these two key things you end up trying to stack a stable pallet without knowing what real estate you are going to need to use next. So the user is left with dealing with one task at a time, leading to a lot of re-stacking halfway through the pick. Sorting of these tasks is your best option to limit the issue, but most techniques can be boiled down to one of two basic algorithms, each with benefits and limitations:

1.       Efficient walk path. 

a.       PRO: You walk the most efficient path possible, therefore reduce the number of steps required to get from A to B.

b.      CON: Without any consideration of individual SKU order quantity you may end up with a pallet appearing to mimic the Leaning Tower of Pisa

2.       Pallet Stability

a.       PRO: By considering the order quantity of individual SKU’s you get a very solid pallet, therefore reducing the probability of damaged goods.

b.      CON: Given the unpredictable nature of DSD ordering patterns, your picker may need to invest in a new pair of walking shoes as they bounce back and forth from one corner of your warehouse to the other.

You may also be able to use slotting logic of your products based on overall volume to align products with your picking patterns. Or perhaps apply some secondary logic to limit bad combinations of products to reduce the issues with either option above, but you are likely still dealing with at least some of these problems (and possibly reducing the beneficial factors as well). I have helped our customers do similar steps in order to get them through these issues but my concerns with these limitations remained.

Lately when discussing their desires for paperless voice picking with customers I have been addressing a different method of implementing this…a mostly paperless solution with voice confirmation, rather than true “picking.” Instead of feeding out the pick commands over a headset the picker gets his pick ticket, with order lines and quantities sorted in one of the methods listed above in addition to his handheld gear. The picker can then identify which product to use as the pallet base. This is of great significance for DSD accounts because all products are not created equal when it comes to providing a stable base upon to build (or for that matter ability to withstand the weight of products stacked above). Look at the PRO/CON list of this offering compared with above:

PRO: 

1.       Pallet stability is maintained by selecting the most suitable base

2.       Pallet stability is enhanced and re-stacking limited (or avoided) based on seeing subsequent pick lines/quantities

3.       Walk path efficiency maintained, with the possible exception of that base product selection

CON:

1.       Selecting of the first “base” product may contradict walk path efficiency

2.       Paper pick ticket still printed rather than pure paper-free voice solution

In HighJump’s voice confirmation solution, the core validation pieces of voice picking are still maintained, i.e. the scan validation and verbal quantity confirmation. In addition, at the end of the pick the picker is prompted with any products missed from the pallet (the main concern of most operational contacts when hearing of this solution for first time).

I think this combination, while not truly paperless, provides the benefits of voice picking and ability allowing instant human interaction to build the best possible pallet. 

Related Posts
Sideload Truck Deliveries: To build by order…or not?

When I first started implementing warehousing solutions for direct store delivery (DSD) and mobile delivery customers, I had come from a WMS background where picking activities were specific to a customer. The idea of aggregate picking was basically a foreign concept. So when a customer would ask for my insight into which picking method I would suggest, the answer always seemed obvious. My belief that this was the best way seemed to follow the industry “trends” at the time as well.

After gaining more experience, and perhaps more perspective on this issue, I now ask customers to question these points prior to committing to a build by order (or palletized) solution suggestion versus building by package (or aggregate) for their sideload truck deliveries. Given some of the responses I have received I think that the trend of palletizing may be losing some steam too.

Labor Cost

The most basic argument for palletizing versus aggregate is that labor costs more for delivery drivers, so you should try to minimize their effort. At face value this seems straightforward and an obvious reason for palletized order building. However, you should be sure that you are able to reap the benefits of reducing the delivery driver labor. If your driver takes an 8 hour shift to deliver while handling aggregate loads, do they really cut their time down to 7 hours (or less) if you provide them with palletized loads? Compare this with your warehouse staff where you have active management in house and stop the shift at the end of the work (or redeploy to other tasks). This type of control over a delivery workforce is difficult to achieve

Cost of Damage (Damage Prevention)

I have never done any research or statistical analysis on the frequency of damaged product that has been built aggregate versus palletized…but I have listened to a lot of warehouse management voice their concerns! A typical first step to alleviate this damage issue is to wrap customer pallets. This normally addresses the issue, but now you have to account for an additional time & materials component when comparing the costs between aggregate & palletized. 

Vehicle Cube Utilization

This is probably one of the most basic arguments against palletizing. Utilization of vehicle cube is obviously much higher while building aggregate. The utilization argument assumes that there is sales volume to get to that level of capacity, which may or may not be the case…but assuming you do have the available volume, by improving cube utilization you may be able to remove vehicle(s) from the road.

Depending on your customer base you may also have to consider dock access. If your sideload trucks are servicing customers with docks where there is the potential of forklift pallet removal, this is a huge swing factor obviously. 

If after considering these factors you decide building palletized orders by customer is the way to go, then more power to you! However, I hope that you consider some of these points before you make a decision based on the assumption that it is “cheaper.”

If you haven’t checked out voice technology recently, you may want to consider it. There are many other common misconceptions about voice technology in the warehouse, which Chad Collins explained recently. Hardware costs have come down quite a bit in recent years, making voice technology an even better value and making it easier to get started. There are also great benefits of adding voice technology to your operations including greater productivity and increased safety. 

Voice for Warehouse Management
The HighJump WMS warehouse management system offers full integration with Vocollect’s Voice-Directed Distribution®, giving you real-time, voice-enabled communication between distribution center workers and the WMS solution. The solution offers a range of voice-enabled picking options configurable by warehouse, specific pick area or specific employee. Benefits include:

• Improve throughput
• Reduce new employee training time
• Improve safety with hands-free and eyes-free picking
• Leverage configurable, voice-enabled picking options

Sport Chalet Success Story – Vocollect and HighJump SoftwareDownload:

 

Voice for Direct Store Delivery Software
You can use voice recognition software in conjunction with HighJump Load Management to manage picking of pallets. Here’s how it works:

• HighJump Load Management optimizes the load for the vehicle and creates the pick instructions. It passes the pick instructions to the handheld device which converts it to voice commands.
• The picker interacts with the software to record any changes to the picks.
• The completed transactions flow back to HighJump Load Management for completion.

If you're committed to supply chain best practices, be sure to consider how voice technology could transform your operations.

HighJump Software just had another customer go live with HighJump Warehouse Advantage and the Embedded Voice module. This customer is a third party logistics provider who plans to use the improved productivity and accuracy of their voice-enabled WMS Warehouse Management System to help them gain additional clients. Yesterday we released a video success story on Fox Racing who also utilizes embedded voice technology in their operations. See the Fox Racing video. The benefits of voice technology in the warehouse seem well understood by most distribution professionals I speak with. Most are quick to point out improved productivity (takes less time to accomplish tasks), safety (workers are hands and eyes free), and accuracy (complex check digit validations). Yet I often hear several misconceptions about voice technology that prevent adoption. Below is a listing of five common misconceptions I frequently hear about voice technology.

 

1.       I require proprietary hardware solutions for voice.

At one time this was true. However, voice technology providers like HighJump Software’s partner Vocollect, have worked with the leading rugged mobile computer manufactures to provide a solution that runs on a traditional device used in the warehouse. It is now possible to utilize the same device to accomplish both voice processes and traditional RF scanning processes.

 

2.       It does not work in the language spoken by warehouse staff.

With HighJump Software’s voice solution, warehouse workers can speak to the device in any language they choose. The recognition technology will respond to spoken commands in any language. Additionally, HighJump Software’s WMS Warehouse Management System is fully internationalized so warehouse employees can be prompted in any of the languages supported in the system.

 

3.       I don’t need a WMS if I have a voice picking system.

Voice enablement is a standard feature in a best of breed WMS Warehouse Management system. The voice technology allows multiple processes such as picking, cycle counting, and put-away to be automated with voice-based interaction with users. A voice system does not optimize all functions within a warehouse and assign warehouse work in the most optimal way. In addition, a best of breed WMS expands beyond the four walls of the distribution center to integrate with suppliers (supplier enablement/supplier integration) and customers. These extended features are not found in voice enablement technology.

 

4.       Voice picking systems simply voice-enable RF prompts.

Most distribution professionals focus too much on the voice technology and not enough on the process differences between traditional RF prompting and voice prompting. Interacting with a text based interface can be completely different than interacting with a voice system. Workflows often need to be optimized for a voice environment. This often means reducing the number of commands spoken and shortening operator response. There is also a different approach to validation because bar code scanning is often minimized in voice environments. 

 

5.       My distribution center is too small for voice.

Voice technology is not just for the “big boys” any more. Simplified WMS integrations make voice implementation straight forward for the end customer. The emergence of multi-modal hardware solutions also mean that distribution centers with smaller teams can have team members easily move between voice enabled processes and non-voice processes. Voice technology represents a strong ROI for any distribution professional who wishes to improve productivity, accuracy and the safety of warehouse employees.