kaoutar bourgane

Data project to better understand and predict delivery round times, using GPS data and business data from delivery operations. - Large-scale GPS data processing: built and optimised distributed jobs on Databricks with PySpark to clean, transform and analyse large volumes of GPS tracks. - Stop and movement analysis: developed algorithms to detect and classify postal workers' stops from GPS tracks, to better understand each phase of a delivery round. - Spatial analysis and enrichment: matched GPS tracks with business data using spatial joins, to qualify and enrich delivery round data. - Data preparation for predictive models: built the training dataset and created geospatial features to predict operational times with XGBoost, together with the Data Science team. - Maps and visual analysis: created interactive maps to explore delivery rounds, check data quality and analyse results on a map. Tools & technologies: Databricks, PySpark, Python, GeoPandas, Apache Sedona, H3, Delta Lake, SQL, Kepler.gl, OSMnx

Management and improvement of the geospatial data platform of Île-de-France Mobilités (Paris region transport authority), in the DATA department, working with public transport and mobility data. - Geographic data platform management: managed and monitored map services, and checked the quality and availability of geographic data used for transport and mobility. - Large-scale geospatial data processing: developed distributed jobs with Apache Spark (PySpark, Spark SQL) and GeoAnalytics for Microsoft Fabric. - Distributed spatial analysis: ran spatial joins, aggregations, transformations and enrichments to produce ready-to-use data for business teams. - Data architecture: used Microsoft Fabric, OneLake and a Lakehouse architecture to store, process and share geospatial data. Tools & technologies: Microsoft Fabric, Apache Spark, PySpark, OneLake, Lakehouse, GeoAnalytics for Microsoft Fabric, GeoParquet, ArcGIS Pro, ArcGIS Server, SQL, REST API

Deployment of a mapping solution for pedestrian navigation inside Transilien train stations (Paris region). - Geospatial data collection: coordinated field data collection and data integration to map 426 Transilien stations, and loaded the data into a structured database. - Geospatial pipeline automation: set up automated ETL processes for OpenStreetMap data with Python scripts (GeoPandas, Overpass API) and open-source tools (QGIS, JOSM). - Data quality tool in production: supervised the development and deployment of a quality control tool for OSM data, based on automated checks. - Data sharing: shared datasets with internal and external users, and took part in the data strategy of the Transilien Data Office (catalogue, documentation, publication). - Agile teamwork: worked with GIS and data teams in Agile mode (Jira, GitLab, Confluence) to ensure quality and traceability of deliverables. Tools & technologies: OpenStreetMap (Overpass, JOSM), Python (GeoPandas, Pandas), QGIS, Jira, GitLab, SQL, OSM API, data quality tools

Build project: set up a GIS solution for the district heating network, compliant with French regulations for works near underground networks (DT/DICT). - Functional scope: defined the project scope, wrote User Stories from business needs, and ran tests and acceptance in Agile mode. - Solution integration: managed the full integration in the client environment (AWS): testing, pre-production and production. - Data publishing and user management: published project data and managed users with Portal for ArcGIS, ArcGIS Online and ArcGIS Enterprise. Tools: Portal for ArcGIS, ArcGIS Enterprise, ArcGIS Online, PostgreSQL, Agile, TargetProcess, Confluence, Jira, AWS

mplementation of an FME solution to monitor fibre optic network deployment. - Data pipelines with FME Form (Workbench): built complex jobs to extract, transform, validate and load data. - Automation with FME Flow: set up scheduled workflows to monitor, check and run geospatial processes at regular intervals. - Data processing and validation: created quality rules, consistency checks and automatic filters to make fibre deployment data reliable. Tools: FME Form, FME Flow, ArcMap, Python, SQL, Oracle

Mise en place d’une solution SIG dédiée à la gestion du patrimoine RTE • Validation des spécifications fonctionnelles détaillées • Définition des tests et recette fonctionnelle de l’application Easygeo • Suivie des bugs et valider leurs correctifs

Administration d'un SIG pour la gestion de la végétation sur les emprises ferroviaires de la SNCF • Gestion des flux de données : Intégration et administration des flux de données liés aux traitements chimiques dans le SIG, assurant une gestion précise et à jour. • Contrôle de qualité des données : Identification et correction des anomalies et incohérences dans les données pour garantir leur fiabilité. • Automatisation des processus : Développement et mise en place d’un modèle automatique pour la préparation des données d’asservissement des traitements chimiques, optimisant ainsi les opérations. • Communication et collaboration : Coordination efficace de la communication interne et externe autour du projet, assurant une transparence et une compréhension partagées. • Outils utilisés : SQL, HTML, ArcGIS Desktop, Dynmap, MySQL, ModelBuilder.

Etude préalable et détaillée pour la mise en place d’une infrastructure de données géographiques (IDG) à la RATP • Analyse de l’existant et des besoins des utilisateurs • Rédaction des spécifications fonctionnelles détaillées • Etude technique : analyse des produits existants sur le marché