| An Integrated Assessment Model of Automobile Smart Cabin Comfort Based on Weight Optimization |
| Wenjun Liao1,2, Xukang Liu1, Jianjun Yang1,2, Yikang Li1, Xiangqun Liu1, Jinghui Ma1, Hongbo Shi1 |
1School of Automobile and Transportation, Xihua University, Chengdu, 610039, China
2Provincial Engineering Research Center for New Energy Vehicle Intelligent Control and Simulation Test Technology of Sichuan, Sichuan, 610039, China |
|
|
|
Received: December 9, 2024; Revised: January 26, 2025 Accepted: January 28, 2025. Published online: April 18, 2025. |
|
|
|
| ABSTRACT |
|
With the popularity of intelligent automobiles, the comfort of the cockpit has been improved. This paper focuses on the comfort of smart cabin, combines the physical environment factors of the cabin and the level of intelligentization of the cabin, and proposes an integrated assessment method for the comfort of smart cabin. Firstly, to establish an integrated assessment system for automobile smart cabin comfort, the study identified five factors that reflect the comfort of an automobile smart cabin: sound, light, thermal, air quality and intelligentization. Eight experimental vehicles were tested on the road, and experts were invited to evaluate the cabin comfort under various operating conditions. The relationship between each comfort index and its corresponding score was then established and a fitting formula derived. Two objective weights were obtained using the EWM and LightGBM, and game theory was used to combine them with the subjective weights obtained using the FAHP, resulting in two combined weights. In addition, the weights were optimized based on the CV coefficient and the KMO measure. Finally, by combining the optimal weighting scheme and the fitting formula, the integrated evaluation model for the automotive smart cabin was established. The effectiveness and feasibility of the model were verified through field studies. |
| Key Words:
Automotive smart cabin · Overall comfort index · Combination weighting method · Weight optimization · Machine learning |
|
Corresponding Author.
Wenjun Liao , Email. 
Preview 
Full text via DOI 
Download Citation
Print
