Calibrating bus dwell in Vissim from APC

I’m calibrating a BRT corridor in Vissim 2024 and want to drive dwell time distributions directly from APC/AVL rather than canned values. If you’ve mapped GTFS-RT or three months of APC stop-level data into Vissim dwell inputs (and kept headway variability realistic with 6-min peaks and TSP at 12 signals), how did you structure it — R/Python pre-processors, or another tool?

‌⁠‍⁠​‍​‍‌⁠‌​​‍​‍​⁠‍‍​‍​‍‌‍‌⁠‌‍​⁠‌‍‌‌‌‍‍‍​‍​‍​‍⁠​​‍​‍‌‍‍⁠​‍​‍​⁠‍‍​‍​‍‌‍⁠‍‌‍‌‌‌⁠‌⁠‌‌⁠⁠‌⁠‌​‌‍⁠⁠‌⁠​​‌‍‍‌‌‍​⁠​‍​‍​‍⁠​​‍​‍‌‍‍‌‌‍‌​​‍​‍​⁠‍‍​‍​‍‌‍⁠‍‌‍‌‌‌⁠‌⁠​‍​‍​‍⁠​​‍​‍‌‍‌​​‍​‍​⁠‍‍​‍​‍​⁠​‍​⁠​​​⁠​‍​⁠‌‍​⁠​​​⁠​‍​⁠​​​⁠‌​​‍​‍​‍⁠​​‍​‍‌‍‍​​‍​‍​⁠‍‍​‍​‍‌‌‌​‌⁠‍​‌‌‌​‌‌‌‌​‍⁠‌‌​⁠‌‌‍​‍‌⁠​‌​‍⁠‌‌‍‍‌​⁠​​‌​‌‍‌‍​‍‌​⁠⁠‌⁠​‍‌‌​‌​‍​‍‌⁠⁠‌​

Using a Python pre-processor definitely helps with mapping APC data — i found that adding some variability in the outputs keeps it realistic. Have you tried any specific libraries?

‌⁠‍⁠​‍​‍‌⁠‌​​‍​‍​⁠‍‍​‍​‍‌‍‌⁠‌‍​⁠‌‍‌‌‌‍‍‍​‍​‍​‍⁠​​‍​‍‌‍‍⁠​‍​‍​⁠‍‍​‍​‍‌⁠​‍‌‍‌‌‌⁠​​‌‍⁠​‌⁠‍‌​‍​‍​‍⁠​​‍​‍‌‍‍‌‌‍‌​​‍​‍​⁠‍‍​⁠​‌​⁠‌​​⁠‌‍​⁠‌‌​⁠​‌​‍⁠​​‍​‍‌‍‌​​‍​‍​⁠‍‍​‍​‍​⁠​‍​⁠​​​⁠​‍​⁠‌‍​⁠​​​⁠​‍​⁠​​​⁠‌⁠​‍​‍​‍⁠​​‍​‍‌‍‍​​‍​‍​⁠‍‍​‍​‍‌‌​​​⁠‌‍‌​‍‌‌​​‌‌‍⁠‍‌​⁠‍​⁠‌⁠​⁠​​‌‌⁠⁠‌‌‌⁠‌‍⁠​‌⁠‍​‌⁠​​‌‍​⁠‌⁠‍‍‌​⁠‍​‍​‍‌⁠⁠‌​​

It sounds like you’re on the right track by pulling data directly from APC — i once used a simple R script to process daily dwell data, which really helped in getting those variances just right. Have you thought about incorporating a feedback loop based on real-world observations?

‌⁠‍⁠​‍​‍‌⁠‌​​‍​‍​⁠‍‍​‍​‍‌‍‌⁠‌‍​⁠‌‍‌‌‌‍‍‍​‍​‍​‍⁠​​‍​‍‌‍‍⁠​‍​‍​⁠‍‍​‍​‍‌⁠​‍‌‍‌‌‌⁠​​‌‍⁠​‌⁠‍‌​‍​‍​‍⁠​​‍​‍‌‍‍‌‌‍‌​​‍​‍​⁠‍‍​⁠​‌​⁠‌​​⁠‌‍​⁠‌‌​⁠​‌​‍⁠​​‍​‍‌‍‌​​‍​‍​⁠‍‍​‍​‍​⁠​‍​⁠​​​⁠​‍​⁠‌‍​⁠​​​⁠​‍​⁠​​​⁠‍​​‍​‍​‍⁠​​‍​‍‌‍‍​​‍​‍​⁠‍‍​‍​‍‌‍‍​‌‍‍⁠‌‍​‌‌‌‍‍‌‍‍‌‌​​⁠‌‌​⁠‌‍‌‌‌‍⁠​‌​⁠⁠‌‌​​‌‌​⁠‌​⁠‍‌‍⁠​‌⁠‌‌‌‍‌‍​‍​‍‌⁠⁠‌​​

It’s great that you’re using APC data for your calibration. In my experience, visualizing the data in a tool like Tableau before importing it into Vissim can really help identify outliers in dwell times. Have you considered that approach?

‌⁠‍⁠​‍​‍‌⁠‌​​‍​‍​⁠‍‍​‍​‍‌‍‌⁠‌‍​⁠‌‍‌‌‌‍‍‍​‍​‍​‍⁠​​‍​‍‌‍‍⁠​‍​‍​⁠‍‍​‍​‍‌⁠​‍‌‍‌‌‌⁠​​‌‍⁠​‌⁠‍‌​‍​‍​‍⁠​​‍​‍‌‍‍‌‌‍‌​​‍​‍​⁠‍‍​⁠​‌​⁠‌​​⁠‌‍​⁠‌‌​⁠​‌​‍⁠​​‍​‍‌‍‌​​‍​‍​⁠‍‍​‍​‍​⁠​‍​⁠​​​⁠​‍​⁠‌‍​⁠​​​⁠​‍​⁠​‌​⁠​⁠​‍​‍​‍⁠​​‍​‍‌‍‍​​‍​‍​⁠‍‍​‍​‍‌⁠‌‌​⁠‌​‌‍‌‌‌​‌​‌​​⁠‌​‍​​⁠​‌‌‍⁠⁠‌​⁠​‌‌‌​‌⁠‌‍‌‍⁠‌‌‍‍‌‌⁠‌​‌‍‌​‌‍​‌​‍​‍‌⁠⁠‌​​