System Thinking: Navigating Old UI/UX in Environmental Modeling

My job involves studying air quality and noise consequences for infrastructure development. Our team uses specialized software from the EPA and FHWA for assessments. However, these government-developed tools often have outdated UI/UX, making the modeling input process error-prone and labor-intensive. For instance, in traffic noise modeling, elements like roadways, elevations, receptors, and obstacles must first be developed in CAD or GIS. Then, we extract coordinates for each geometric element, and then we manually input traffic values.

System thinking is a holistic view of processes, combined with automation, has revolutionized our workflow. We can automate input processes during off-hours using Python and third-party packages like PyAutoGUI, reducing manual errors. This approach was applied to one of our noise projects, where we had 12 weeks to model the noise impacts of a new highway. The project involved digitizing 1,582 roadways, assessing noise impacts on 2,383 receptors, and examining 27 barrier systems.

Without system thinking, the team would manually input each of the 1,582 roadways, determine elevations, and assign the traffic/speed for each roadway. This manual approach risks errors, such as missing a roadway, misassigning traffic data, or inconsistent input methods among team members. A system thinker would use a programmatic approach to minimize potential errors. Once modeling inputs were developed, errors would mainly come from the source data, reducing mistakes during the input process.

Applying system thinking to the project meant ensuring accurate preparation of all input elements. We aimed to represent all modeling components in as few datasets as possible. This required preparing various shape files before the input process, including the following:

? Point file for roadway vertices that contain roadway name, XYZ coordinate of the roadways, etc.

? Receptors file that contains data such as land use, address, existing noise level, build noise level, proposed mitigation barrier name, and efficacy of the barrier

? Barrier files containing each of the proposed barriers that include the height and whether the barrier is proposed based on the DOT criteria.

? Traffic file that contains the roadway name and the vehicle volume for each vehicle classification

The systematic approach emphasizes automation where possible and rigorous quality control where automation isn't feasible. For instance, roadway naming is crucial as it links roadway input to traffic data. Quality control ensures no duplicates in unique identifiers and that traffic data aligns with the line diagrams from the traffic study outputs. Of the 12-week project timeline, the first 8 weeks were dedicated to preparing inputs. For the remaining 4 weeks, 2 weeks were spent on automation, receptor input, debugging the input, and implementing other noise-specific components such as ground zone and flow control on roadways. The last 2 weeks were spent identifying impact and coming up with the 27 different barrier systems along the project, then digitizing the barrier and running the height variation to determine which barrier would be feasible and reasonable for the project.

With the introduction of automation, workflows inevitably evolve. Our experience with the project underscored that the bulk of our time is now channeled into preparatory work. This shift prompts us to question how to maximize automation during the 16 off-hours, allowing us to focus on refining and debugging models during active work hours.

Interestingly, applying system thinking can sometimes challenge traditional methods. For instance, in the project, no modeling activities occurred for 8 out of 12 weeks of the project duration. This efficiency, while beneficial, might impact the billability of engineers skilled in manual model input.

To navigate this new landscape, I propose the following strategies:

1. Top-Down Approach: The increased efficiency from automation can free mid-management from routine project tasks, allowing them to explore growth opportunities. This shift necessitates reevaluating mid-management's utility goals, aligning them with the overarching objective of serving more clients with enhanced quality.

2. Bottom-Up Approach: The leadership team should foster a platform for system thinkers within the organization. This forum would enable discussions on tasks ripe for automation, ensuring continuous workflow and quality improvement of products.

#SystemThinking #Automation #Python #UI/UX #TrafficNoise #ChangeManagement #EnvironmentalAssessment #GIS #CAD