Visual Asset Management System
- File uploading process
A centralized database to store and manage large 3D assets.
Project Overview
What is VAMS?
Visual Asset Management System (VAMS) is a centralized database to store and manage large sets of refinery/equipment reality capturing, metrology, and 3D model data to enable the digital twin initiative.
Business Goals
Centralized database to reduce duplicated projects
Users can seamlessly manage, find, and re-use data
Help project teams find the best quality vendors for capturing projects
Impact
~$200k cost savings per refinery by reducing duplicate projects
Designed the uploading process that increased data uploading speed by ~43%.
Decreased system data caching time by ~77% by re-organizing and reducing metadata schema.
Designed a role-based access system to cater to different data access tiers for better asset management.
Digital Twins
My Responbilities
Throughout my internship, I worked on 3 core functions of VAMS: project creation and file uploading process, role-based access for data and system operation, and metadata-driven searching and filtering.
This case study will focus on project creation and file uploading process.
Team – I worked with internal stakeholders and a 3rd party development team
My Role – UX designer /researcher
Project length – 3 months
Tool – Mural, Adobe XD, pen and paper
Visual design guide – Company’s internal standard design system
Not a fan of research?
Project Timeline
There were 3 Sprint cycles I went through for this project. Each cycle had different goals. I shifted direction twice and adapted to new project challenges and requests.
First Sprint
Since the project team had a tight schedule to put MVP in production, the initial launch would only include reality-capturing projects. Therefore, I focused on collecting information from the most direct users – 3rd party scanning vendors who upload most of the reality-capturing data.
Research
Stakeholder Interview
Instead of going right into the users, I interviewed my product manager and 2 core project members to understand their goals, needs, and requests for the project.
Key Requirements
User Interview
With project requirements in mind, I started preparing my user interviews. One challenge was identifying and finding users since they were third-party vendors with whom I had limited contact. To tackle this, I contacted all site project managers and core project members for a list of vendor companies and individual vendors they worked with. From there, I selected three vendors with different experience levels and conducted interviews.
Design Goals
Competitive Analysis
Partial selected direct and in-direct competitors
Design
User Story Mapping
Key user pains
"How might we optimize the uploading process to
easily allow users to attach relevant project metadata
and monitor the overall uploading progress?”
After user interviews, I did a competitive analysis to see how existing products tackle the uploading of large sets of visual and 3D assets and manage location-based data.
Key learnings from the competitors
After consolidating user problems, project requests, and learnings from existing software, I focused on problems that can be solved on the front end through design. Before sketching, I created user story map to show user flow and functions that would address user pain points and stakeholder requirements for the project team to review.
Design Solutions
After aligning with the project team, I created sketches and wireframes based on the user story map.
Second Sprint
Emerged problem
The second sprint happened before my first round of usability testing. The project scope expanded from just reality capturing to including metrology scans. Metrology scans are used for drastic different projects and users.
Additional Research
User Interviews
To fill the gaps in my initial research for not including internal engineers who work with metrology grade scan, I interviewed two engineers who leverage metrology scan data differently to have broader perspectives on user needs/pain points.
Key User Pains
Most of the data are stored on local hard drives or stand-alone computers. Therefore, searching and filtering files is difficult. Users rely on their file naming system to organize data.
1st Usability Testing
Design
Key findings
Real projects will have more than one scan type. Currently, workflow only allows users to stick to one scan type per project.
Some users don’t understand file types associated with file buckets while others do not categorize their data the same way.
Unlike reality-capturing scans, metrology scans are sometimes used as individual files without attaching to a project.
Metrology scans require different metadata fields for identification and search functions besides location information.
The clickable breadcrumb is a better solution than the file tree. Users used words such as “familiar” and “used to” when using breadcrumbs, indicating an established mental model. However, users completed tasks without any assistance on both designs.
I relayed the test findings about the file bucket to the project team to see if the naming conventions could be changed in a more user-friendly manner. The project team explained the rationale behind the file types and how they will become more critical once the system is used at a larger scale. I made the file type an optional dropdown instead of mandatory buckets to mitigate user confusion and project requirements.
Iterated User Story Map
Iterated Design
Third Sprint
By the time the second sprint was over, the project team had to adapt to the company’s data team request for integration with the company’s internal database, which will push through 70 metadata fields into the VAMS. I communicated with the project team that we would need to reduce metadata fields and only select ones useful for file search and management. Moreover, reduced metadata fields will help the development team to minimize data caching time.
As more teams within the company are interested in leveraging the digital twin technology, 3D model engineers want to leverage VAMS to better share data with other engineers. I set up interviews and usability tests with 3D model engineers to understand their work process, evaluate how scalable my design is, and potentially modify my design to solve their pain points.
Additional Research
Open Card Sort
To reduce and categorize the metadata fields for the uploading process and project display page, I did a card sort activity with 5 different users who work on different types of projects. Besides card sorting, I also observed the users’ file naming conventions to identify the critical metadata through recordings of previous user interviews.
One card sort example - Orange cards are the frequently used metadata. Pink cards indicate the metadata that does not exist within the internal database but is important to users.
User Interview
2nd Iteration
Key learnings
Most users must use a handful of certain metadata combinations to identify the information. Having the ability to save and reuse those fields is essential.
Most metadata fields have pre-established relationships. For example, site names to country, geographic location, and region. Therefore, only showing the most helpful field - ‘site names’ to users and keeping the less valuable related fields ‘ geographic location’ and ‘Region’ in the backend will reduce time spent selecting metadata fields during uploading.
I interviewed the 3D model engineer to find out more about their work process, pains, and needs for data management.
Key User Pains
Large quantities of data are attached to 3D models. Users sometimes lose data during the transfer process.
Users need to update parts frequently, on average every 2-3 weeks. Reuploading three terabyte models every 2-3 weeks becomes a huge pain.
3D model visualization tool relies on the folder hierarchy. One small change in folder hierarchy in the database may corrupt the model.
2nd Usability Testing
To validate my iteration and identify design gaps with 3D model and metrology engineers, I conducted the second round of usability testing.
Potential Long-Term Solutions
Key findings
Clickable breadcrumb is not clear for 3D model engineers to identify file hierarchy, which might break the 3D model in real-life projects.
The 3D model engineer wants the system to remember which file needs to be updated regularly.
Due to the limited time left in my internship, I couldn’t research further and design my long-term solution to address the needs of 3D model engineers. However, I provided the immediate solution – bring back the file tree to display to provide better visibility for folder hierarchy.
Users could leverage workflow to pre-select what files need to be updated regularly. When the updated files are uploaded, the system will place them at the right place automatically.
User journey or user history map that allows users to click on the folder they went to without going back to the file display to find the correct folders repeatedly.
Prototype
2 ways to review selected file to suffice different use cases for different scans
Clickable Breadcrumb
or File Tree (Showing folder hierarchy)
Clear overall uploading process
Locate uploaded files
locate and reupload failed files
Simple flow to a new project
Select location
Draw a bounding box of the exact scan location on a map,
Attach scan types (Modality) per file
Keep irrelevant metadata fields hidden to optimize the system loading time
Upload stand-alone files without attaching to a project
Learnings & Reflections
Initially, I thought this would be a simple task. As my research and testing go further, I realize the level of complexity of an essential function of software.
Much going back and forth between research and design in a real-life project. However, if I were approaching this project again, I would do two things differently to make the process more efficient.
First, I would gather information on all types of users, categorize them, and time interviews and research more strategically to prevent multiple changes in the design flow and system functionalities.
Instead of focusing on individual functionality, I would think about how the entire system can help users complete their significant tasks to make the system more scalable.
Lastly, it’s essential to have a 2-way system to bring any usability issues caused by back-end issues to the engineering team. Samely, back-end issues should surface and find potential solutions through UX design.
