Sibyl: Predictive Maintenance for Industrial Washing Machines

Industrial washing machines are crucial for sanitation in hospitals and hotels, but maintenance specialists are few and far between.

Project Specs

Timeline: 10 weeks

Deliverable: product concept

My role: design researcher, systems designer, visual designer, illustrator

Skills: IBM enterprise design thinking, user research, journey mapping, storyboarding, storytelling

Tools: Adobe Illustrator

Team: Lindsey Thompson, Will Kuglen, Tess Horany, & Katherine Kytka

Special thanks to our mentors: Brooks Protzmann, Doug Powell, & Devin O’Bryan

How might we streamline the repair process for industrial washing machine maintenance specialists and reduce pain for the machine facility owners?

“You want us to look at washing machines?”

On the first day of UT’s inaugural classroom partnership with IBM, we were shuttled to IBM’s Austin campus and told we would be receiving a design challenge. Each of us was placed in an interdisciplinary team, and each team was given a topic. My team was assigned … industrial washing machines. Huh?

As it turns out, maintenance issues for industrial washing machines are common, but maintenance specialists in Central Texas are scarce. This leads to loss of time and money, overworked employees, late deliveries, and the risk of losing customers. We had 10 weeks to research the problem and propose a solution, meanwhile following IBM’s Enterprise Design Thinking.

Championing Our User

A site visit to a laundromat propelled our project by giving us a user to champion. We interviewed “Owen the Owner” who gave us critical information about three problems with his business.

Maintenance records for Owen’s machines were not digitized, but instead kept on a note card inside the machine.
In the event of a breakdown, Owen might not know the machine isn’t working until several days later, since he is busy managing multiple locations. When he does learn of a breakdown, he schedules a maintenance visit, which could take up to several weeks. Even then, the specialist might not have the needed part on hand, resulting in more machine downtime. It’s not uncommon for Owen to tip his specialists hundreds of dollars to keep a good relationship in hopes of his repairs ending up at the top of the list.
In the worst case scenario, undetected machine malfunction can cause a fire to burn down the facility, and this actually happened to one of Owen’s former locations.

An Engine Light for Your Washing Machine

To address Owen’s pain points, we decided to propose a predictive maintenance approach. What if our solution could be like your car’s engine light: it lets you know about breakdowns before they happen, so you don’t end up stranded. Sensors could be installed in the machines, connected to the IBM Cloud, pull data from the machines, and send it to IBM Watson for interpretation and next steps.

Imagining the Interface

With our deadline fast approaching, we planned to present an interface that demonstrated the main capabilities of our concept: prediction, purchasing, and scheduling.

Screen Shot 2020-07-03 at 12.39.18 AM.png

Screen Shot 2020-07-03 at 12.39.26 AM.png

Screen Shot 2020-07-03 at 12.39.34 AM.png

Our screens communicated the point, but not the story. Quickly, we pivoted away from designing the interface and instead decided to put our resources toward telling a compelling story about our user. Foregoing a robust UI allowed us to focus on conveying the emotional appeal of our user’s journey to our audience.

I used Adobe Illustrator to create a set of custom illustrations to tell our story.

We named our product concept Sibyl after the oracles of Ancient Greece because of its predictive capabilities.

The Big Day

At the end of the sprint, our team took on the colossal challenge of making washing machines sound exciting to our audience. As well as presenting to people who were along for the ride all semester, we delivered our story to some who were hearing it for the first time. Our strategy was to express the problem as a story through the lens of Owen the Owner.

By presenting our as-is and to-be journeys back-to-back, we were able to highlight Owen’s pain when things go wrong and compare it to the relief he feels when Sibyl is in play. In the end, we didn’t need a product mockup to tell a good story or get buy-in from our audience.

As-Is User Journey

This Owen, the owner of Laundry 101. Owen has several facilities in the Austin area. Today, we will be focusing on his downtown location.

At this location, owen has specialized commercial washers and dryers and a 1 folding machine.

He also has 12 employees and 80 clients. Facilities like Owen’s typically service small business like clinics and or hotel spas.

One Monday, Owen pops in on his downtown location and sees that two machines are out of order.

Unable to diagnose the problem himself, Owen calls Martin, his maintenance specialist. Martin is the only guy in the area who can service machines like Owen’s.

It’s Monday, but Owen can’t get Martin to come until Friday. This means his employees are going to have to work overtime to make up for two machines on downtime all week.

When Martin arrives, he identifies a broken belt in Owen’s machines.

But Martin doesn’t have the particular part on hand. He’s going to have to order it and come back later.

Martin takes off, leaving Owen frustrated and still waiting.

Owen has a big problem on his hands.

A week later, Martin returns with the part in tow.

He’s finally able to service the broken machines.

Owen’s machines are good to go!

But Owen is burdened by a huge bill at the end of this painful process.

And rather than feeling relieved, now he feels like the rest of his machines are just breakdowns waiting to happen.

To-Be User Journey

When we last saw Owen, he was left feeling anxious because he still has 34 washing machines that could break down at any moment.

That’s where Sibyl comes in.

Owen has IBM Ingrid come to his facility and hook up sensors in his machines to the IBM Cloud.

Now, Owen and Martin can use Sibyl on their phones to communicate about maintenance issues.

Now, all of Owen’s machines are connected to the cloud, enabling data to flow to existing technologies, like Owen’s smartphone and computer.

Okay, so now we’re going to walk through the same problem Owen faced in his As-Is, but now utilizing Sibyl’s capabilities.

Owen gets an alert on his smartphone: “Hey Owen! …”

Owen can see exactly which 2 machines at which location need maintenance attention. These machines are still currently running, but need action to keep them that way.

Sibyl shows Owen the part at risk.

Ahead of time, Owen can order a part by choosing from a list of vendors.

Owen can see Martin’s availability and select a day when the part will arrive.

Martin sees Owen’s request and knows everything will be ready when he gets there.

The parts are waiting for Martin by the time he comes to Owen’s facility.

Martin can do his work quickly…

And in no time, Owen’s machines are good to go.

Owen is thrilled by the cash he gets to keep in his pocket this time.

That’s a big difference!

With Sibyl, Owen feels secure in his business and his machines. Martin’s time is reduced to only one visit. IBM Ingrid was happy to empower people like Owen. Owen’s workers aren’t overworked, and his clients feel confident that their laundry will be delivered on time.

Sibyl helps Owen and Martin cut out costly steps and get the job done right the first time.

Caroline Rock

UX Researcher