2528. Optimizing Air Pollution Sensor Networks: Integrating Stationary and Mobile Sensors Under Budget and Calibration Constraints
Invited abstract in session MA-55: Network Optimization 1, stream Network Optimization.
Monday, 8:30-10:00Room: Liberty 1.09
Authors (first author is the speaker)
| 1. | Mohammad Jafari
|
| 2. | Bart van Arem
|
| Delft University of Technology | |
| 3. | Serge Hoogendoorn
|
| Transport & Planning, Delft University of Technology | |
| 4. | Maaike Snelder
|
| Civil Engineering, Delft University of Technology |
Abstract
Air pollution has become a significant global challenge, affecting public health, ecosystems, and economic stability. Accurate monitoring is essential for informed decision-making. While high-cost stationary sensors provide precise data, their limited deployment reduces overall coverage. Low-cost mobile sensors (LCSs) mounted on buses can complement stationary sensors, improving both spatial and temporal coverage. However, LCSs suffer from poor accuracy and require frequent calibration through single-hop (near a stationary sensor) or multi-hop (between two mobile sensors) methods. A gap exists in integrating sensor coverage and calibration in mixed stationary and mobile sensor networks under budget constraints. This study formulates a non-linear optimization model to select the optimal subset of buses and sensor station locations based on real-time bus location data and spatiotemporal grid weights. The objective is to maximize sensor network coverage while considering budget limitations and ensuring that all mobile sensors are calibrated frequently. The results of the model, applied to a Rotterdam case study, show that with 11 buses and 2 stationary sensors, 50% of the grids can be covered. The model output demonstrates that optimizing bus and sensor station selection enhances both calibration and coverage efficiency. Further studies can explore utilizing a mix of mobile platforms to enhance sensor network coverage and improve calibration efficiency.
Keywords
- Network Design
- Programming, Nonlinear
- Transportation
Status: accepted
Back to the list of papers