Scholarlink

This study investigates the continuity of electricity distribution service in mainland Portugal using quantile regression models applied to two key technical indicators: TIEPI (Time of Interruption Equivalent to Installed Power) and END (Energy Not Delivered). Based on data from E-REDES (2014–2022), the analysis evaluates how interruption frequency and duration, measured by SAIFI (System Interruption Average Frequency Index) and SAIDI (System Interruption Average Duration Index), affect energy supply resilience across different service quality zones and municipalities. Results show that SAIDI has a consistently positive and increasing effect on both END and TIEPI across quantiles, especially at the uppermost, indicating higher exposure to supply loss in the Municipality Codes with poor performance. Fixed effects highlight considerable municipal level heterogeneity, while QSR zone classifications alone do not significantly explain performance after accounting for local variation. These findings support the need for geographically target investment and regulation, and demonstrate the value of distribution sensitive modeling in infrastructure performance assessment.

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

Modeling the Service Quality Indicators TIEPI and END Using Quantile Regression

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Marina A. Andrade,
  • M. Filomena Teodoro

摘要

This study investigates the continuity of electricity distribution service in mainland Portugal using quantile regression models applied to two key technical indicators: TIEPI (Time of Interruption Equivalent to Installed Power) and END (Energy Not Delivered). Based on data from E-REDES (2014–2022), the analysis evaluates how interruption frequency and duration, measured by SAIFI (System Interruption Average Frequency Index) and SAIDI (System Interruption Average Duration Index), affect energy supply resilience across different service quality zones and municipalities. Results show that SAIDI has a consistently positive and increasing effect on both END and TIEPI across quantiles, especially at the uppermost, indicating higher exposure to supply loss in the Municipality Codes with poor performance. Fixed effects highlight considerable municipal level heterogeneity, while QSR zone classifications alone do not significantly explain performance after accounting for local variation. These findings support the need for geographically target investment and regulation, and demonstrate the value of distribution sensitive modeling in infrastructure performance assessment.