Nicholas Fournier, PhD

Refactor of core travel data pipeline (statistical weighting). Reduced project-level of effort from weeks to hours. Eliminated bottleneck with configurable pipeline allowing analyst level staff to operate. Reduced risk by implementing automation, CI/CD testing, linting, and QAQC reporting and validation.

Wrote high-performance python API for concurrent requests for self-hosted routing machine (OSRM) for bulk routing and map-matching of GPS traces.

Contributed core model to open source activity-based travel model ActivitySim for a disaggregated accessibility measure estimator, linking higher level models to lower level choice models.

Developed toll pricing “futures” market model using bi-parabolic macroscopic traffic flow model and price elasticities to optimize traffic flow. Funded by California State SB1.

Developed graph theory-based bicycle network connectivity performance measure. Funded by Caltrans.

Python program to detect operational but mislabeled traffic sensors using variety of machine learning techniques (e.g., k-Nearest Neighbor, Logistic Regression, Random Forest, Support Vector Machines, Local Outlier Factor, Isolation Forest, and Robust Covariance Anomaly Detection). Funded by Caltrans.

UX research/human factors study to determine bicycle infrastructure preferences using virtual reality bicycle simulator. Estimated using a Latent Class Choice Model capable of accounting for user heterogeneity. Funded by Caltrans.

Advised team of 19 doctoral students conducting industry partnered research in public transportation engineering, planning, policy analysis, and economics.

Generated synthetic population of Boston for activity-based travel simulation model as part of joint MIT/UMass ARPA-e competitive research project. Developed a novel data fusion method to synthesize fixed work location data in population synthesis.

Conducted human factors research using full-scale immersive driving simulator with eye-tracking and control monitoring (steering, gas, and brake) to study responses to novel roadway infrastructure. Sponsored by the U.S. DOT SaferSim UTC.

Developed a sinusoidal model for seasonal bicycle demand estimation for calculating bicycle-vehicle crash risk where bicycle traffic data are limited.

Evaluated and prepared a report on the approval process of roadside safety hardware for the Federal Highway Administration.

Collected and analyzed LIDAR and video data from instrumented vehicle to estimate microscopic car-following model parameters.