Prompt

Each day in the Market Delivery Operations Warehouse, associates work to prepare shipments that contain bulk items as well as individual pieces of pipes and other types of conduit. Since the creation of the MDO warehouse, the process of accounting for the individual pieces of pipe has been a manual one. My team was tasked with creating a digital flow for piece picking that ensures quick and accurate picking, as well as providing warehouse managers with the ability to track piece bundles as they move throughout the facility .

This manual process is time consuming and leads to countless mistakes. For example, wrong number of items picked, wrong items picked, bundles assigned to the wrong shipments, etc.

Research

The research plan consisted of observations as well as user interviews. See below for a breakdown of the research demographics.

• 1 UX’er (Facilitator & Observer)

• 1 PM (Observer)

• 5 General Warehouse Associates/Piece Pickers (3 months-5 years exp),

• 3 Managers

• 3 Separate Warehouses

• 2 States  (Georgia and Texas) 

Research Findings

• On average, it takes about 5-10 minutes to pick and label each bundle.

• Shipments can have multiple piece pick bundles. Usually no more than 3 bundles. Some shipments have no bundles.

• If there is a requested piece that happens to be out of stock, the picker must manually write that on the corresponding pick list printout they received at the beginning of their shift. The front office is notified once all printouts have been returned to the front office at the end of the shift.

• Overall, each pice picker felt a digital solution would be helpful. 

• However, the more experienced pickers did not feel that a digital solution was necessary because it could possibly slow them down.

• The managers agreed that a digital solution could possibly slow down the more experienced pickers. 

• However, they felt it was more important to be able to track how pieces move through the warehouse and ensure the right pieces end up with the right shipments.

Solution

After completing our research, it was time to get back to the office and and share our findings with our engineering team. After talking through some things and a couple of white boarding sessions to hash out some technical stuff we were ready to move into the designing phase. Our findings from research told us that our proposed solution needed to:

• Allow piece pickers to pick quickly 

• Allow piece pickers to pick accurately

• Allow piece pickers to update out of stock items in real time

• Allow piece pickers to easily associate bundles with the correct shipments

• Allow managers to track how inventory moves through the warehouse

*Prior to conducing research we learned that most of the warehouse associates rely on an Android handheld radio frequency gun that is used for picking bulk items. That being said, the solution for piece picking needs to work on the RF gun as well. 

In this first iteration, we knew that the warehouse associates would still have to rely on the pick list paper mentioned above in the Research Findings section. At the moment, there were some technical restraints keeping us from eliminating the paper, so we’re keeping it for now. In this flow the user begins by scanning the barcode on the pick list then, they officially start picking. Next, they head to the location of the item they want to pick and scan the location label. If there is enough inventory to fulfill the request they continue by confirming the actual quantity without changing anything, if not, they adjust the quantity and then confirm. The location scan and confirming of the actual quantity steps are repeated until all available items on the pick list have been picked. Once the system registers that all available items have been picked, it’s time to confirm that all pieces have been picked. Once this happens the warehouse associate will assign the bundle(s) to either a staging location or with the actual shipment. If it goes to a staging location, that just means the loader who is collecting the bulk items for the assigned shipment will need to pick up the bundle(s) when they are done loading bulk items to the flatbed truck.

User Testing

The team (PM and UX) flew back to Texas to conduct some user testing sessions . Our goal was to test our flow with at least 5 warehouse associates. We ended up testing with 6 users. For the tests, I created a prototype using Invsion. We came up with a list of assumptions to test against. Doing this helps us determine how well (or not) the sessions went. See below for some additional info and findings.

Test Overview

• Testing Device: Android Mobile Device

• Participants: We tested with 6 end users

• Scenario: Imagine your next shipment includes 2 bundles of product. Using this device please walk us through how you would pick and then stage or load the shipment.

Findings

• None of the users were able to determine that they should scan the barcode on the pick list in order to begin the picking flow.

• All users were able to adjust and confirm the actual quantity successfully

• 4 out of 6 users were able to load or stage the bundle(s) successfully

• 5 of the users specifically asked to see the pick list on the RF device rather than on paper

• 4 out of 6 users expect to make less mistakes with a digital flow.

Overall, the warehouse associates were able to move through the flow with ease after realizing they would need to physically scan the paper. We were in a good spot to head back to Atlanta and get the engineering team started on coding the screens.

At this point, I officially handed off the screens to dev. Our product team shared the same work area so anytime a design related question came up I was there to talk through it with them.

Getting Launched

On launch day, our team went back to the MDO in Georgia. While there, I answered questions and facilitated a few mini training sessions with warehouse associates and managers to ensure they felt confident navigating the digital flow.

Early Wins

Some immediate benefits we observed post-launch:

• Real Time Updates: Associates and Managers appreciated the ability to mark items as out-of-stock on the spot, reducing reconciliation errors.

• Increased Visibility: Managers gained real time visibility into where piece bundles were located, facilitating improved coordination between departments.

• Faster Picking Times: 12 associates expressed that the digital workflow was faster than their previous process after using it for one full shift. Reporting showed decreased picking times even on launch day, showing an average of shipments being picked 35 seconds faster than before launch.

  • Prior to launch, it took anywhere from 11-16 minutes to pick bundles for a shipment.

  • After 1 month, reporting across all MDO warehouses showed shipments were being picked and staged or loaded an average of 2 minutes faster than before launch. This increased the number of shipments pickers were able to complete in a shift. Now, the Piece pickers were picking 5-10 shipments more per shift than previously.

Software Used

• Sketch

• Invision

• Mobzien (Screen recording app)