This study aims to empirically verify the applicability of a Driver Advisory System (DAS) for reducing energy consumption in electric multiple units under actual revenue-service operating conditions. Previous studies on railway energy saving have ...
This study aims to empirically verify the applicability of a Driver Advisory System (DAS) for reducing energy consumption in electric multiple units under actual revenue-service operating conditions. Previous studies on railway energy saving have mainly focused on vehicle performance improvement, infrastructure enhancement, regenerative braking utilization, and Automatic Train Operation (ATO). However, relatively limited empirical research has been conducted on energy consumption variation associated with driver operation behavior under actual railway operating conditions. Therefore, this study analyzed the operating characteristics of electric multiple units on the Gyeongchun Line and examined the relationship between energy consumption variation driver-operation-related factors based on actual operation data. and The target section of this study was defined as the Sinnae-Namchuncheon section of the Gyeongchun Line. The Sangbong-Sinnae and Namchuncheon–Chuncheon sections were excluded because frequent signal-related waiting and variations in operating conditions made quantitative energy comparison unsuitable. First, long-term operational data from electric multiple units operated without DAS were used as baseline data to analyze actual speed distance profiles, power running, coasting, braking patterns, and section-specific operating characteristics. These non-DAS operational data were not used as direct validation data for the S-DAS application effect, but as fundamental data for generating economic driving profiles and identifying baseline operating patterns under actual revenue-service conditions. Separate S-DAS trial operation data and revenue-service monitoring data for non-S-DAS and S-DAS operations were then used to verify the tracking performance of the economic driving profile and the observed energy-saving effect of S-DAS. Based on the analysis of baseline operating patterns, track conditions, vehicle performance, driving constraints, and actual operation data, this study designed a Standalone Driver Advisory System (S-DAS) with a non-interfering, on-board independent structure. The proposed S-DAS does not directly intervene in the existing traction, braking, or signalling control systems of the vehicle. Instead, it uses position, speed, energy consumption, traction notch, and braking information collected from the Train General Information System (TGIS). The system compares the current operating state with the economic driving profile and provides advisory information to the driver through a Human–Machine Interface (HMI). In addition, an Ethernet-based data interface structure was applied to improve real-time performance, scalability, and maintainability. The applicability verification of S-DAS was conducted using operation data obtained from the same vehicle system and interpreted under the same TGIS-based data processing criteria. The validation results using S-DAS trial operation and revenue-service monitoring data showed that the calculated reference energy consumption based on the economic driving profile for a round trip between Sinnae and Namchuncheon was 692.74 kWh, while the actual S-DAS trial operation energy consumption was 687.55 kWh. The deviation from the reference profile was −0.75%, indicating that the economic driving profile generated in this study can function as a reference profile that can be practically followed in actual electric multiple unit operation. In the revenue-service monitoring results, the average energy consumption decreased from 781.07 kWh in non-S-DAS operation to 725.75 kWh in S-DAS operation. This corresponds to an average reduction of 55.32 kWh. Based on the non-S-DAS average energy consumption, the average energy consumption reduction rate was approximately 7.08%. In addition, when the same energy difference is interpreted relative to the economic-driving reference energy consumption of 692.74 kWh, it corresponds to an observed excess-energy reduction effect of approximately 7.98% relative to the reference profile. Therefore, the 7.98% value in this study should be interpreted separately from the average energy consumption reduction rate. The excess-over-reference rate decreased from 12.75% in non-S-DAS operation to 4.77% in S-DAS operation, corresponding to an improvement of 7.98 percentage points. The similarity rate between the actual operation results and the economic driving profile increased from 87.25% to 95.23%, also corresponding to an improvement of 7.98 percentage points. These results indicate that, after S-DAS application, the actual operation results moved closer to the economic driving profile. In this study, the 7.98 percentage-point value represents the improvement in the excess-over-reference rate and the similarity rate, rather than the average energy consumption reduction rate. However, the comparison between non-S-DAS and S-DAS operations was conducted under actual revenue-service conditions. External factors such as operating date, signal conditions, passenger load, train spacing, dwell time variation, and driver operation differences could not be fully controlled. Therefore, the results of this study should not be interpreted as an absolute or universally generalizable reduction rate obtained from a fully controlled experimental test. Instead, they should be interpreted as observed changes in average energy consumption, excess-over-reference rate, and economic-driving-profile similarity under the same route, section, and TGIS-based data processing criteria. In addition, the interpretation of driver operation behavior in this study is based on outcome indicators such as energy consumption, excess-over-reference rate, similarity rate, and economic driving profile tracking performance, rather than on a direct and separate evaluation of individual driver control actions. Accordingly, this study does not claim direct causal identification of individual driver behavior. Rather, it presents an empirical interpretation of the relationship between energy consumption variation and driver-operation-related factors under actual revenue-service conditions. This study is significant in that it interprets railway energy consumption not only from the viewpoint of vehicle and track conditions, but also from the perspective of a human–machine integrated system involving driver operation behavior and driver advisory functions. It also empirically demonstrates the applicability and observed energy-saving effect of DAS through actual on-board application, trial operation, and revenue-service monitoring data. The findings of this study are expected to serve as basic data for expanding DAS application to general electric multiple units, improving driver advisory systems, and establishing railway energy-efficiency policies. Keywords: Driver Advisory System, S-DAS, Electric Multiple Unit, Economic Driving Profile, Energy Saving, Driver Operation Behavior, Applicability, Empirical Study