模擬捷運轉乘站候車月台的人際接觸時空分布：利用智慧卡資料與代理人模式

Abstract

流行病容易在密閉且擁擠的大眾運輸系統上傳播，其中轉乘站更是高風險區域，因為乘客在這裡會與來自各地的人群接觸，進而提高感染風險。而傳播擁擠資訊是一種降低感染風險的方式，我們已經可以透過網路地圖或運營商提供的擁擠資訊來避開高風險車廂或車站，但這些資訊未能呈現更詳細的站內人際接觸特性。 先前的研究已利用智慧卡資料探索乘客在車廂內的共存人數，並將其視為潛在接觸人數，然而對於車站空間的接觸特性，目前仍未得到充分的討論。因此，本研究的目的為探索乘客在轉乘站月台人際接觸特性的時空分布，為達成此目的，本研究首先將智慧卡資料結合時間地理學 (Time geography) 的概念，估算乘客在不同時段於轉乘站月台的共存人數與擁擠密度，以初步評估接觸特性。之外，為了更詳細探索乘客在月台的接觸特性，本研究進一步運用代理人基模型 (Agent-based Model, ABM) 來模擬乘客候車位置的空間分布以計算近距離接觸人數，並透過現場觀察來確認模型的有效性。綜合上述，本研究的方法有助於更詳細的了解捷運轉乘站月台的人際接觸特性，並可以作為大眾運輸運營商實行擁擠管理或是公共衛生機關實施防疫措施時的參考。

Epidemics tend to spread easily on enclosed and crowded public transportation systems, with transfer stations being particularly high-risk areas. This is because passengers at these locations meet people from all over, thereby increasing the risk of infection. One way to reduce this risk is through the dissemination of congestion information. We can already avoid high-risk areas by using online maps or operators' congestion information. However, these resources do not provide detailed information about the interpersonal contact characteristics within the stations. Previous studies have used smart card data to analyze the co-occurrence of passengers in carriages as potential contacts. However, the characteristics of contact within station spaces have not been extensively discussed. Hence, the aim of this study is to explore the spatio-temporal distribution of interpersonal contact characteristics on transfer station platforms. To achieve this aim, the study merges smart card data with the concept of Time Geography, estimating the number and density of passengers on transfer platforms at different times. This serves as a preliminary assessment of contact characteristics. Furthermore, to delve into the contact characteristics on platforms more thoroughly, this study utilizes an Agent-based Model (ABM) to simulate the spatial distribution of passenger waiting positions, calculating the number of close contacts. The effectiveness of the model is then confirmed through field observations. In conclusion, this study's method enhances the understanding of interpersonal contact characteristics on subway transfer platforms. It serves as a guide for transit operators in managing congestion and health agencies in implementing preventive measures.