The goal of the project was to ease the activities of the delivery drivers for the vaccine delivery logistics process. The initial process involved a lot of paper handling and lacked real-time information at the control center. We intended to bypass conventional paper-based reporting systems and submit reports instantly using an application on their mobile device.
Our high-level goals were to:
- Make it easy for the drivers to know their daily activities
- Manage Routes efficiently
- Manage delivery commodity quantities and variance
Logistic Management Information Systems (LoMIS) is a suite of offline-capable mobile and web applications that address challenges with the supply/distribution of health commodities to last-mile health facilities.
The LoMIS Suite applications provide critical solutions to ensure the availability of health commodities, effective vaccine management, and enable broader health system policy decisions through the use of real data for insight and execution at the community, state, and national level.
For the sake of this case study, our focus will be on the last-mile health facility deliveries by the drivers, which resulted in the design of the LOMIS Deliver Mobile App.
Vaccine Delivery EcoSystem
- Drivers are responsible for picking up the planned health facility’s allocations from the cold store and delivering them to the designated health facilities. They play a major part in the process therefore it is essential to improve their experience and help them perform jobs faster and more efficiently.
- State Cold Store is the usual start of the project scope, this is the location where all vaccines from the national repository are stored for any given state. The usual pickup point for the daily delivery route.
- Project Coordinator/Vendor are the assigned fleet/logistic/program manager designated to run the process for any given state. They own and coordinate the drivers and vehicles. Also manages the activities and planning of the exercise
- Health Facility is the last-mile location where the deliveries are supplied. Usually, a mini-cold store to serve the eventual recipients of the vaccines
There are existing systems to manage most arms of the ecosystem and a web application used by the vendor to assign, distribute and optimize routes. Our focus again is the challenges of drivers.
Understanding the user contexts and needs
To understand the challenges we needed to understand the current contexts on when/where/how the users interact with the process and the current challenges faced.
Note: a hybrid digitization process was in place but yielded little result due to low participation/corporation of the drivers. Using a mix of Google sheets and paper forms.
Using qualitative, unmoderated research approach with emphasis on the diary study method. we were on the field with the drivers for a couple of weeks gathering pictures, recording activities, and logging pain points.
Below is a summary/contextual analysis from the driver’s activities and interviews
By going through the various contextual situations the following problems were prioritized for the first iteration of the application
- Knowing daily assigned tasks
- Computing the number of commodities to pickup from cold store
- Computing and reconciling delivery commodities
- Managing and recording accepted and delivered quantities
- Efficiently mapping routing from planned destinations
- Getting feedback to the command center and continuity of work at low or offline network regions
Sketches and fleshing out the ideas
Did a quick sketch to see how the journey will interact and look
The sketches were converted to wireframes to validate the flow with the product team. A major focus was placed on selecting a task and entering delivery data.
High Fidelity Prototype
We had a few drivers come into the office to gather feedback on the prototype, but most of the responses were positive. So we scheduled a field visit using the current prototypes alongside the current process.
I was surprised to see a lot of issues popping up during this exercise
- The drivers needed to know which of the tasks were synced because the coordinator was consistently calling to find out the status of every delivery
- Deliveries are not necessarily made on designated days, some activities could be pulled forward if there were resource capacity Note: the deliveries of these days are moved in totality.
- The coordinator notifies them on a regular of changes or events surrounding their deliveries
- Visibility under natural or sunny light was essential
- Entering values on the input using the keypad can be very challenging
Hi-fidelity Mockups (Second Iteration)
Eventual Updates in the second Iteration
- Higher contrast colors
- Bigger value input boxes
- Sync indicator on every task item
- Easy switching between dates and indicators to help identify relevant ones.
- Using a counter for value entry (This also helped with counting products)
Conclusions and Reflection
The design of the Lomis Deliver App had a tremendous impact on the delivery service, increasing the rate of delivery due to timely and effective plan changes. Driver interactions and usage also saw a significant increase with positive feedbacks on the experience.
Moderated assessment or validation may not yield the true result in every case, going to the field with the drivers using the prototypes really showed the true impact of things.
Like every projects there always a few hinderances, for Lomis deliver they were surprisingly minimal. The organisation understood the need for proper UX and made available a lot of the resources (Time) needed to get proper feedback and research. But, the field workers were initially reluctant to adapt a mobile technology solution. Luckily carrying them along through the development process made them understand they were building for themselves by themselves.
- 91% uptime of cold chain equipment in Kano state in 2019
- 14,083 successful deliveries of vaccines and dry goods were tracked using LoMIS Deliver, in Bauchi, Sokoto, and Zamfara state
Note: Critical parts of the system were left out to avoid leaking trade secrets and voiding NDAs