In my Interaction Design studio course, our group was tasked to design a service-based mobile app for American Machine and Foundry (AMF) Bakery Systems. We delved into the world of factory equipment and designed for the machine operator as a user, making sure to have a logical user flow and microinteractions to enhance the experience.
As 4 college students, none of us had any knowledge in the industrial factory equipment arena, so we commenced heavy research on the domain.
AMF lists among its clients each of the largest baking companies in the United States as well as baking companies in Germany, New Zealand, Mexico, England, Puerto Rico, Trinidad, Saudi Arabia, Canada, Poland and Sweden. (source)
So if you’re eating a hamburger from McDonald’s, a pretzel, a muffin, or just slices of bread, it was probably produced with an AMF machine!
“We believe that success comes from an expert understanding of the complete bread, bun and soft roll baking process. Our commitment to complete solution excellence, backed by innovative technologies, is what makes us unique. It is what makes us AMF.”
The AMF Vision
Here we could see that AMF was very accepting of new technologies in order to better the company, so we were excited to see what we could make to make operations run even more smoothly.
Before reading this, you might not have known what AMF bakery was, but now you can see how vital it is to the Big Mac and other staples in America. So naturally there are many stakeholders with needs and goals for AMF.
want AMF to be the leader in the industry,
need to bring on new clients and keep current clients satisfied to create a trust and dependency, and
prefer to decrease costs on maintenance and repairs of equipment.
Buyers, such as McDonalds and Flower Foods Inc
want keep their brand culturally relevant,
need consistency in the quality of their food, and
prefer long lasting equipment with little to no downtime.
Consumers of these baked goods
want, need, and prefer good food at a good price.
We kept all of of this in mind as we marched on with our research.
Concept Map & Value Flow
We mapped out major concepts related to AMF and determined that there were three key areas of operations for the company: products, marketing, and distribution. We also identified relevant relationships and interactions between these areas of interest.
This helped with the next step of understanding the problem space: a value map
We took the concept map and narrowed down the scope of the project to some key areas of the business. Here we found that there were potential bottlenecks in maintenance requests, as machine operators were the ones that processed them, which can be time consuming, and can slow operations down due to dependencies on both the factory equipment and the operator. So now we had a specific user journey we could work with.
Introducing Jerry, a machine operator at a factory we'll call Dunder Bread.
Whenever there’s a repair needed for a machine, he has to talk to Maya, the buyer representative. She then has to take time out of her day to work on requesting a repair.
This indirect process can result in a large amount of downtime for the machine, and a huge loss of revenue for the client. Jerry is waiting for the repair, Maya feels overwhelmed by the additional responsibility.
Now that we understood the current situation, our team was ready to start designing.
After identifying our realm to work in, we wrote down our problem: A heavy reliance on the operator causes a bottleneck in operations
From here we wrote down 20 questions to help push us to design an effective product. We ended up bringing up interesting points like:
How do we access the operator’s knowledge?
What are the operator’s key challenges in his/her daily routine?
These helped us to start ideating what services a machine operator requires.
Next we used reverse assumptions to challenge the status quo of machine operators at AMF.
One key assumption we had was that operators manage maintained and repairs, but what if they didn’t? We played with this idea in our prototyping stage.
From that reverse assumption, we focused on the machine repair process to see how a mobile service app could mitigate the work of an operator.
In our low fi wireframes, we captured the general process of the requesting for repair process to get an overall feel of what it looked like. But it wasn’t really that different from an operator manually filling out a request form to AMF, so we kept pushing to see what else we could come up with.
After some critiques from others, we pivoted a bit to focus on the repair process of the machine itself. In our mid fi wireframes. We came up with the idea of giving the operator access to the status of machines on the factory floor, and allowed him/her to scan a machine to see what could be wrong with it and allow requests to be sent easily afterwards.
We then narrowed even further in our high fidelity screens to just scanning a machine.
The idea we had was that the operator would simply just capture the machine in its state, and the application could take in the implicit data, such as the age of the equipment, run time, and so on, as well as explicit data, such as the number of previous repairs of the machine and the video capture by the operator to generate a preliminary diagnosis of the machine to send to AMF.
By adding a micro-location feature, we were able to design for operators’ busy work days on factory floors, during which they are unlikely to check their phones. The notification feature prompts the operator to check specific machine depending on their location on the factory floors.
By recognizing the cost associated with that call, we developed a tiered repair recommendation system. The predicted percentage for likelihood of an issue is dependent on explicit and implicit data collection.
In the “machine capture” process microinteractions here, operators get both visual and haptic feedback when recording the correct parts of a machine.
The operator maneuvers his camera to align the overlay on the screen with the corresponding machine part. Once the app recognizes the alignment, it “locks” into place and begins video and audio recording.
After recording, operators receive a recommendation. We worked to incorporate visual forms of data-driven insights and reasoning for operators on this screen
The Future & Impact
Let’s see how Jerry and Maya fare with the new system in place. Now that Jerry can capture the machine status on his phone, he can directly send it to AMF along with the diagnostic from the data. Jerry no longer has to take time to reach out to Maya, and Maya no longer has to worry about repair requests for the machine.
Our team predicts this solution has the potential to be machine learning process, so that when the application gathers enough explicit and implicit data from clients, it can accurately asses a machine’s health through existing data rather than a manual capture process.
Designing for a big company which involves stakeholders across different divisions and striking the balance between business strategy, client needs and customer is challenging but rewarding. Getting more specific to the needs of their particular industry and exploring pain points & opportunities during the design process helped us think holistically and critically.
Given the limited amount of time and opportunity space, we learned how to narrow down our scope and how to frame our business idea to bring the maximum value for our stakeholders. We also explored how we could bring value that was unique to our mobile digital touchpoint.
Design is not only focusing on how to create delightful experience, but more importantly, how we can use design solutions to incorporate into people’s daily lives and streamline their day-to-day routines.
Special thanks to Neha Shah, Lily Pai, Tiffany Liu for being great team members to work with!