Belib Spatial Analytics ← Back to portfolio
Geospatial & Algorithmic

Belib Location Optimizer: Multi-Layer EV Analytics

An optimization engine for mesh and geospatial allocation developed to analyze the saturation of electric vehicle charging infrastructure (EVSE) in Paris. The pipeline maps, discretizes, and evaluates the topology of 1,200 urban grids by cross-referencing real-time traffic flows and geodetic proximity.

Author: Avner EL BAZ
Frameworks: GeoPandas, Shapely, Public API Data Paris, Python

Coverage Gain

+28.4%

Optimization of access points within 400 meters.

Stations Modeled

1,200+

EV charging points indexed dynamically via public API.

Intersection Speed

<14ms

Per polygon query using R-Tree indexing.

Cross-Join Complexity

\(\mathcal{O}(N \log N)\)

Instead of \(\mathcal{O}(N^2)\) via Haversine matrix optimization.

Dissatisfaction Index Analysis by District

Spatial Algorithms & Geodetic Modeling

To avoid plane projection distortions at the intra-muros Parisian scale, the pipeline converts topographic coordinates via the Haversine geodetic formula. This establishes an exact distance matrix between traffic centroids \(C_i\) and charging station coordinates \(S_j\):

d = 2R \cdot \arcsin\left(\sqrt{\sin^2\left(\frac{\Delta \phi}{2}\right) + \cos(\phi_1)\cos(\phi_2)\sin^2\left(\frac{\Delta \lambda}{2}\right)}\right)

To define optimal locations (congestion index \(I_c\)), the model combines normalized traffic flow intensity \(F_n\) and the inverse density of the existing charging network within a buffer radius \(r\). The objective function aims to minimize the residual variance of this distribution:

\(I_c(x) = \sum_{i=1}^{N} \frac{F_n(C_i)}{\text{dist}(x, S_i)^2 + \epsilon} \cdot \mathbf{1}_{\{\text{dist}(x, S_i) \le r\}}\)

Analysis Report

View the full cartographic and statistical results of the urban study.

PDF Preview (Expand on wide screens or download the document)

Download Full Report (PDF)