臺灣二手電動車市場行銷策略分析

Abstract

中文摘要 近年來，隨著電動車市場快速成長，二手電動車逐漸成為消費者與車商 關注的新興市場。然而，二手電動車價格受品牌、車齡、里程數與電池壽命 等多重因素影響，市場資訊不對稱問題明顯，增加交易風險與決策難度。 本研究以臺灣二手電動車市場資料為研究對象，透過 Python 進行資料 清理與探索性資料分析，檢視價格分布、品牌結構及價格與里程數、年份之 關聯性。進一步運用相關係數分析與品牌分層策略，說明品牌溢價對價格判 讀之影響。 在此基礎上，本研究建構隨機森林迴歸模型作為 AI 估價工具，並透過 特徵重要性分析，驗證品牌與里程數為影響價格之關鍵因素。最後，結合模 型結果發展智慧採購判斷、庫存折舊分析、行銷文案生成與 Web 應用系統，將資料分析成果轉化為具體可執行之行銷與決策支援工具，提升二手電動車市場之決策效率與實務應用價值。 Abstract With the rapid growth of the electric vehicle (EV) market, used electric vehicles have become an emerging segment attracting increasing attention from both consumers and dealers. However, pricing in the used EV market is influenced by multiple factors, including brand, vehicle age, mileage, and battery condition, leading to significant information asymmetry and higher transaction risks. This study analyzes Taiwan’s used electric vehicle market using Python-based data science methods. Exploratory data analysis is conducted to examine price distributions, brand structures, and the relationships between price, mileage, and manufacturing year. Correlation analysis and brand stratification are further applied to reveal the impact of brand premium on price interpretation. Based on these findings, a Random Forest regression model is developed as an AI-based pricing tool. Feature importance analysis confirms that brand and mileage are the most influential factors in price prediction. Finally, the model is extended to practical applications, including intelligent purchasing evaluation, depreciation-based inventory strategies, automated marketing copy generation, and a web-based pricing system. The results demonstrate how data-driven and AI-powered approaches can support marketing strategy formulation and decision-making efficiency in the used EV market.