探討後疫情時代下的遠距活動行為：以混合選擇模式為應用

本研究旨在探討近期廣泛被採用遠距活動對個人參與實體活動的影響，並分析社會影響在個人決策中的作用，以及其如何影響個體的選擇行為與旅運模式。以台灣六都為範疇，調查遠距活動的選擇行為及影響因素，為更真實反映選擇行為的變化，本研究設計遠距購物選擇情境的敘述性偏好問卷，以三種購物模式：店內購物、自取點取貨和宅配到家為替選方案，方案屬性包括產品與運輸配送，以構建遠距購物模式與重要屬性變數。結果顯示受訪者多數偏好自取點取貨方案，可能與台灣自取點較密集有關。遠距活動影響實體旅運，涵蓋替代性、互補性、調整性與中立性，隨疫情趨緩影響力減弱，而其中調整性最顯著，受訪者認為能靈活安排行程，其次為替代性。店內購物和自取點取貨方案之重要屬性為交通往返時間，而改善產品折數可提升宅配到家方案的市佔率。個人態度對遠距購物行為的選擇偏好中，順從性與永續性是影響消費者選擇購物與配送方式的重要因素，並在多項羅吉特模式證實，納入潛在變數後可有效提升解釋能力，在本研究之混合羅吉特模式中，誤差成分項分析發現，自取點取貨在交通運輸層面與店內購物相似，在線上購物模式上則與宅配到家相似。此外，受訪者對產品折扣與交通往返時間的偏好存在隨機異質性。實務策略方面，以各方案進行模擬，發現擴大產品範圍5%等同產品折扣增加5%，可藉由增加自取點密度以降低旅行時間，可相對提升市佔率約3.76%，多品項集運提供產品折數5%，即等同於消費者接受多等待一天之效果。

This study aims to investigate the impact of the recent widespread adoption of telepresence activities on individuals' participation in physical activities and to analyze the role of social influence in individuals' decision-making and how it affects their choice behavior and travel patterns. To more truly reflect the changes in choice behavior, this study designed a stated preference questionnaire for the online shopping scenario using three alternatives: in-store, click & pick, and home delivery, and the attributes of the choice scenarios included product and delivery perspectives. It can construct the choice model of online shopping and delivery behavior, and identify those important explanatory variables. The results show that major respondents prefer click & pick, which may be related to the high concentration of pickup points in metropolitan area. The influence of remote activities on physical travel covers substitutability, complementarity, adaptability and neutrality, which diminishes as the epidemic subsides, with adaptability being the most significant, as respondents perceive it to organize their trips flexibly, followed by substitutability. The most important attribute for in-store and click & pick is the travel time while improving product discounts could increase the market share of home delivery. Conformity and sustainability are the most important factors affecting consumers' attitudinal preference for online shopping and delivery behavior, and the inclusion of latent variables can improve explanatory power, as confirmed in the result of Multinomial logit model. Additionally, the specification of error component on Mixed logit model reveals that click & pick is similar to in-store in terms of transportation but similar to home delivery in terms of online shopping. In addition, there is random heterogeneity in respondents' preferences for product discounts and transportation round-trip time. In terms of practical strategies, the result of proposed strategy reveals that expanding the product range by 5% is equivalent to a 5% increase in product discounts, increasing the density of pickup points can reduce travel time, which can relatively increase the market share by about 3.76%. Finally, a sensitivity analysis of the sustainability strategy for consolidated delivery shows that a day of slow delivery only provides a 5% discount on products.