Scholarlink

As electric bikes gain widespread popularity across Indonesia, efficient and interoperable battery-swapping stations are increasingly necessary. This study uses machine learning, specifically an LSTM neural network, to predict the interoperability of these stations by analyzing demographic and geographic data, such as population density, urbanization rates, and electric motorcycle distribution. The automated process also considers proximity to major transport hubs, age, and income distribution. The LSTM model achieved 92% accuracy, significantly surpassing conventional models, which typically reach around 80%. Cities, with their younger, tech-savvy populations and higher adoption rates of electric motorcycles, present better opportunities for system integration. These findings offer crucial insights into how battery-swapping stations can be optimally deployed and operated. By addressing challenges such as interoperability, this research aids stakeholders in building a reliable, efficient, and user-friendly battery-swapping network. Additionally, it contributes to strategic policy-making for electric vehicle infrastructure, promoting sustainable transportation and supporting future urban planning efforts in Indonesia.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Predicting Interoperability at Battery Swapping Stations for Electric Motorcycles in Indonesia: A Geographical, Demographic, and LSTM-Based Machine Learning Approach

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Feri Adriyanto,
  • Wahyudi Sutopo

摘要

As electric bikes gain widespread popularity across Indonesia, efficient and interoperable battery-swapping stations are increasingly necessary. This study uses machine learning, specifically an LSTM neural network, to predict the interoperability of these stations by analyzing demographic and geographic data, such as population density, urbanization rates, and electric motorcycle distribution. The automated process also considers proximity to major transport hubs, age, and income distribution. The LSTM model achieved 92% accuracy, significantly surpassing conventional models, which typically reach around 80%. Cities, with their younger, tech-savvy populations and higher adoption rates of electric motorcycles, present better opportunities for system integration. These findings offer crucial insights into how battery-swapping stations can be optimally deployed and operated. By addressing challenges such as interoperability, this research aids stakeholders in building a reliable, efficient, and user-friendly battery-swapping network. Additionally, it contributes to strategic policy-making for electric vehicle infrastructure, promoting sustainable transportation and supporting future urban planning efforts in Indonesia.