As the global temperature continues to rise and concerns over greenhouse gas emissions escalate, the imperative for adopting greener technologies has never been more pronounced. The shift towards electric mobility stands as a promising strategy to decarbonize the transport sector, with many nations rallying behind initiatives like the global EV30@30 campaign, which aims to see at least 30% of new vehicle sales being electric by 2030. According to McKinsey’s projections, by 2030, approximately 120 million electric vehicles (EVs) could populate the roads across China, the European Union, and the United States.
A fundamental prerequisite for realizing this ambitious transition lies in establishing an accessible and robust network of EV charging infrastructure. To support this, numerous countries have enacted enabling policies aimed at fostering the development of charging infrastructure and empowering stakeholders to facilitate its expansion on the ground. However, the efficient and timely implementation of EV charging infrastructure necessitates a contextual approach that aligns with local requirements and integrates seamlessly within existing electricity supply and transportation networks.
One of the primary challenges is around ensuring the alignment between the readiness of charging stations and the actual demand. Site selection for optimal deployment of charging stations is a multifaceted process. It involves considerations such as electrical service availability, traffic density, travel hours, land ownership, and crucially, social equity. These diverse factors are synthesized to estimate demand and provision services efficiently, thereby maximizing the utilization of charging stations and minimizing carbon footprint.
The landscape of EV charging infrastructure includes various charging equipment, each with its charging time, battery capacity, and compatibility with different types of electric vehicles. Charging stations are typically categorized based on AC/DC charging levels and serve different purposes. This is not limited to public access charging, workplace charging, and DC fast charging. Advising customers on selecting a specific charging station involves optimizing constraints such as charging speed and vehicle compatibility.
The integration of Geographic Information Systems (GIS) with GeoAI and Machine Learning algorithms provides a comprehensive platform for synthesizing these constraints. By analyzing demographic data, customer visit patterns, and spatial characteristics, these technologies facilitate the identification of potential customer clusters and the evaluation of site suitability for charging stations tailored to specific vehicle categories.
An array of factors are considered in demand analytics to conduct Site Suitability Analysis for Charging Stations. Parameters assessed are demographic data, societal trends, vehicle types, and tariff structures to estimate demand across various categories of Charging Stations. Employing multivariate analysis techniques, including Weigh-Rank Models and Design of Experiment, aids in identifying potential demand clusters. Also, Support Vector Machine and Logistic Regression methods yield promising results in delineating these clusters.
The identified customer clusters, along with parameters from site suitability analysis, are pivotal in evaluating optimal locations for Charging Stations. This process encompasses Multi-criteria Multi-objective Decision Making, where factors such as Charging Station category, emission reduction through optimal distance alignment, and time efficiency are key objectives. It also involves assessing the utilization efficiency of existing stations and considering the replacement or displacement of outdated sites. Given the constraints of limited cruise range and recharge times, innovative routing algorithms that prioritize energy efficiency are imperative. By integrating these constraints, an optimal energy-efficient route is derived, thereby curbing emissions.
The outcomes of this analysis include:
- Mapping Charging Stations with Customer Clusters based on customer and Charging Station categories, distinguishing between public and private facilities.
- Establishing vehicle-type associations with Charging Stations.
- Mapping public or private stations while considering the Social Equity aspect.
The end-to-end solution in EV Charging Station Suitability Analysis encompasses six blocks of analysis, covering vehicle categories, Charging Station types, and the Effective Utilization Index, which measures coverage efficiency for customer clusters.
The end-to-end solution follows a structured execution pipeline outlined as follows:
- Data Acquisition: Primary datasets sourced from Ground Truth or Satellite Imagery, along with Secondary datasets, are obtained either online or offline and then fed into the pipeline.
- Pre-Processing: Each dataset undergoes geocoding, interpolation or extrapolation as necessary, normalization, and scaling to prepare them for Demand Analytics (customer clusters) and site suitability analysis.
- Customer Cluster Identification: Utilizing the GIS platform, the solution delineates customer clusters based on vehicle categories and land use clusters (residential, commercial, industrial, etc.) to determine the appropriate category of Charging Station (Public Access Charging, Workplace Charging, DC Fast Charging).
- Objective Definition: Each execution within the pipeline commences by defining specific objectives and concludes by outlining how the results can be utilized to pinpoint potential locations for further investigation.
- Multi-Criteria Multi-Objective Analysis: Various multivariate optimization algorithms, such as the Huff Model and others, are employed to extract potential Charging Station sites. These sites are selected based on the identified Customer-clusters and an Attractive Index determined by the objectives set in the model.
With the rising adoption of electric vehicles and expanding consumer demand for charging, the task of strategically locating charging stations becomes increasingly intricate. The need to stay within the network while adhering to ESG guidelines adds another layer of complexity. With the introduction of vehicle GPS and the availability of communication base station data, more and more studies are using real-time EV trajectory data and location data to generate real-time demand generation for a vehicle type, and customer profile to map with a Charging Station optimally in terms of ETA, Optimal Distance aligned with Reduced Emission and help prepare a Scheduler for Charging in real-time.
