A Human-Friendly Production Management System for Cross-Functional Teams

ORCA is a large-format touchscreen first user interface for managing apparel production. While there are existing ERP and PLM packages for this industry, they are all either difficult to use or exceedingly expensive. Billion-dollar brands like GAP build and maintain their own internal applications to support the design to manufacture to retail process, but nothing like that exists for small companies operating $5 million in revenue or less. Given the cost and high technical barrier, many fashion brands don’t adopt an ERP or PLM system until they are over $10 million in revenue and it becomes inoperable to not do so.

The challenge lay in integrating user needs from a variety of different stakeholders. In smaller companies, people often will perform multiple roles. Also, the long term benefits of creating an easy to use interface lay in enabling the designers to more readily manage their own production, instead of needing to hire more production managers. I talked with many team members to gain insight on what their processes and needs are, especially paying attention to any current workflow steps that resulted in errors or double-checking.

Existing Systems

The production lead had set up a card-based system of tracking styles. The cards group the style with the fabric options offered for that style, as well as some high level information like product code and due date. They used post-it notes and highlighters to track production status as well as exception cases. Significantly, the board of cards is ever-present and highly visible to everyone on the team, a big improvement over a spreadsheet or folder of POs. 

This contrasted greatly with the current version of the software build that had been in progress for 6 months. The app captured a lot of details, but the screens were scant improvement over a spreadsheet, and much of the UI was boilerplate forms. 

I proposed that the board of cards should become the natural basis for the UI, with a large touchscreen that could replace the analog board so that everyone could work from the same platform. The team quickly agreed. The team desired to keep many of the paper card’s functions that they had established processes around, so much design iteration revolved around how similar the UI should be to the physical cards while integrating within a digital platform.

Version 0

I made a quick prototype of the cards to test how they might feel on a screen. This led to quick agreement about the overall direction.

Version 1

I focused on how inputting information would work in collaborative sessions when the team is planning a product line. A standard form input is very linearly tedious. The benefit of physical paper on a board is being able to move around things all at once. A designer doesn’t think about a fashion collection one item at a time, but rather thinks about what styles to include, and then what fabrics to match to each style. So, I designed this drag and drop interface so that you can move styles and fabrics on multiple cards all at once.

Since this is tricky to imagine in wireframes, I made a prototype to test it out. Avoiding the need to have a whole database of items, I simply used images of styles and fabrics from a folder in Finder.

Version 2

I sought to represent the various status updates on the cards in a layout where the colors would be visible from a distance, but wouldn’t be too distracting. 

Tab organization between more screens emerged, with a screen for tracking Fabric and another tracking individual products. These replace a massive and confusing spreadsheet that the team currently used.

Versions 3-5

The team really wanted to keep the highlighting and post-it note metaphors for tracking production status. So I did several iterations with feedback to define how many types of notes and statuses there needed to be. While skeumorphic UI elements have become outmoded in consumer apps, the team really wanted the notes to feel like post-its. Since we are transitioning a currently paper-based system to digital, this is a valid intermediate step. Also the post-it approach does add depth to a touch interface with high density of information.

I didn’t focus on making everything visually perfect, but rather on capturing all the variations of user needs. The board both “zooms in” to show more details for one style, as well as “zooms out” to display two collections side by side.