Ellis Don
SmartStruct is a mobile application solution, created for Ellis Don, that leverages data and AI to act as a smart assistant for construction site supervisors.
-
Problem
It is important to report problems quickly so that delays are mitigated. Most crews rely on verbal or written instructions. Walk arounds occur frequently to make sure the work is being done and issues can be flagged. Some construction apps are currently on the market, such as Procore, PlanGrid, Raken, and Fieldwire. Slow adoption is common since crews have become accustomed to current practices. However, better reporting needs to occur so that in-office staff are aware of project progress. Delays and inflated costs are common concerns. Therefore, there is an opportunity to mitigate unexpected issues on site so that personnel can determine how to change the plan and prevent costly delays.
-
Challenge
How might we help site supervisors keep daily work moving forward efficiently when unexpected issues arise so that major project delays are avoided or minimized?
Project details
-
Duration
5 days
-
Role
UX/UI Designer
-
Team
Rhea D’Souza
Denise Bentum
Victoria Rudiak
Approach
-
Research: secondary, competitive & interviews
-
Persona: goals, motivations & pain points
-
Inspiration & ideation
-
Sketching then wireframing
-
Task flows & information architecture
-
Prototyping & presenting
Research
Research goals
Learn about the current methods of tracking daily productivity and providing progress updates
Identify key players
Learn about industry trends to consider for solution
Competitor analysis
Process
-
98% of large-scale construction projects suffer cost overruns of > 30 %
77% are at least 40% late
-
Poor short-term planning and missed connection to actual progress. A close examination of root-causes shows that more than 80% of performance variations are related to changes in work plan, poor communication of resource availability, incompletion of handoffs among subcontractors, and underestimating effort required to complete a task.
Insufficient management of project risk. Current metrics do not offer any insight about risk for potential problems and the reliability of a project’s look-ahead schedule. Project teams are not equipped with actionable analytics that allows them to proactively tap off potential delays before they surface on their project sites.
-
Things go wrong every single day
Site supervisors are largely responsible for capturing daily productivity and providing updates on the progress of the project
Workers are annoyed/frustrated when their day and work is impacted by the progress of other trades on the job
-
The construction apps we looked at for functional and visual inspiration included Autodesk Bim 360, Fieldwire, PlanGrid, and Raken. Tasks should be accompanied by images, data visualization, automated weather reporting, simplistic and minimalist design patterns were considered in the construction of Smartstruct.
Assumptions
Construction workers have a smartphone and use it regularly
Work is project specific
The project is long term
All crews have foremen and site supervisors on site
For construction workers, it’s not an efficient use of their time to do this reporting. The foremen and site supervisors are receiving work updates, so they should input the data.
Currently, this kind of reporting is being done either verbally or with pen and paper.
Subcontractors and employees all report progress to foreman – whether internal or external.
User persona
Jack Howes, Site supervisor
Goals: Meet project deadlines & ensure quality work.
Pain point: Unexpected issues arise each day leading to wasted time & project delays
“All the dominoes have to fall into place or it can throw off the rest of the timeline.”
Inspiration
Inspiration was collected from Dribble & project management websites. Focusing on the data visualization, gantt chart layouts, and simplistic visual identities, we created a visual language for Smartstruct.
Ideation
From the ideation process, we decided on 3 main components that we wanted the solution to have:
-
From our research, we learned that risk is very important to consider in construction.
The app would use various data points and lessons learned to assign risk levels to tasks
-
When a problem arises, the app will generate a series of required tasks and smart solutions that allow the site supervisor to efficiently reassign crews to the new task and adjust the schedules accordingly
-
After the Site Supervisor approves and implements the Smart Solutions, the schedules and plans are automatically updated.
The Super will be able to see how the project timeline (plan) is affected by specific daily activities and problems.
Initial Sketches
Task flow
Information architecture
We identified three main areas for our app: Dashboard, Alerts, and Plans. Next, we organized our content within these categories, figuring out where our main ideas fit.
-
Snapshot of the day including:
High risk tasks
Weather
Date
Personel
Recent BIM updates
-
Problem alert includes:
Details of problem
New required tasks (smart solution for trade/crew assignment and evaluating problem impact)
-
Project timeline includes:
List of tasks including those in progress
Includes risk, trade, crew location, deadline and auto updates
Wireframes
High fidelity prototype
Reflection
-
Impact: Creating a smart assistant for supervisors
Smartstruct leverages data and AI to help site supervisors reduce inefficiencies by assigning tasks a risk level, suggesting smart solutions when problems arise, and automatically adjusting the project timeline accordingly.
EllisDon workers are already collecting a lot of data through applications such as Fieldwire, but not much is being done to harness the potential that all this information could provide. Smartstruct would learn from past projects and lessons learned to provide the most efficient and appropriate solutions.
Ideally, Smartstruct would be used at multiple EllisDon construction sites. This would not only make the app smarter by giving it more data to work with, but also allow materials and resources to be shared between sites if needed.
-
Key learnings & recognition
Research is pivotal when the subject matter is unfamiliar. Through interviews, we were able to gain insights into who the key players were and the recurring problems that occur on site. Very little resources and information was provided and the problem space presented was quite open-ended.
Taking initiative to go on site and talk with the people who the product would be effecting was important in gaining empathy with the user and allowing our application to be successful in the eyes of Ellis Don. Smartstruct was named runner-up in the Ellis Don hackathon.
-
Future thinking: Facing the resistance to technology
Implementing mobile-based software can be difficult in a somewhat-traditional industry like construction, where there is often much resistance to the use of computer technology. The regular tradesman may not recognize the potential value, or may simply see it as not worth the effort. Smartstruct may face similar issues in becoming widely adopted, but by targeting the Site Supervisor specifically, the process may not be too difficult. There are a lot fewer Site Supervisors, and they would be able to see the direct benefits from using the app. There is not a lot of work for them to do, and it actually builds on an existing app that many of them already use.
One large consideration is that the effectiveness of Smartstruct lies in the data that is being collected, thus, the data collected from Fieldwire must be complete and of quality. If the foremen or tradesmen are not motivated enough to record good data, consistently, then Smartstruct may fail to meet the needs of the Site Supervisors. However, mindsets are changing as the construction industry looks to software and Big Data to improve its processes. Adoption will only get easier as mobile-based technology use becomes more commonplace on the construction site.