The city of Mumbai has grown at an unprecedented rate, increasing the burden of mobility on its core public transport system, the Mumbai suburban railway network. The system is likely failing from “over-optimization,” with stations not designed to cater to the needs of a rapidly growing city, which has led to a steady surge in fatalities over the years, primarily in the metropolitan region beyond the city limits. Besides fatalities, research indicates that crowding has led to extreme fear and insecurity, especially in women and young commuters, with inappropriate behavior by fellow passengers causing them extreme discomfort. There is a need to decongest the Mumbai suburban rail network across the system and to gain a better measure of the extent of crowding in and around transit facilities. Concepts such as level of service (LOS) from the vantage point of crowding science can be used to address this need. However, there are two critical challenges. First, concepts developed in the Global North are inadequate to deal with the kind of commuter densities and complexities typical of cities like Mumbai. Secondly, conventional data gathering methods have proved to be time consuming, costly, and too inflexible to capture dynamic commuter behavior critical to the science of crowd management. This paper aims to address these two challenges by articulating a set of “inquiries” that can inform a localized framework and share learnings from the application of basic video and image processing. Thus, it proposes dynamic data capture methods that inform and enable a scientific planning process.

孟买市以前所未有的速度发展，增加了其核心公共交通系统孟买郊区铁路网的交通负担。该系统可能因“过度优化”而失败，车站的设计无法满足快速发展的城市的需求，这导致多年来死亡人数稳步上升，主要是在城市范围以外的大都市区。除了死亡人数外，研究表明拥挤会导致极度恐惧和不安全感，尤其是女性和年轻的通勤者，其他乘客的不当行为会导致他们极度不适。需要疏散整个系统的孟买郊区铁路网络，并更好地衡量交通设施内外的拥挤程度。从拥挤科学的角度来看，诸如服务水平 (LOS) 等概念可用于解决这一需求。然而，有两个关键挑战。首先，在全球北部开发的概念不足以应对孟买等城市典型的通勤密度和复杂性。其次，传统的数据收集方法已被证明是耗时、昂贵且不灵活的，无法捕捉对人群管理科学至关重要的动态通勤行为。本文旨在通过阐明一组“查询”来解决这两个挑战，这些“查询”可以为本地化框架提供信息并分享从基本视频和图像处理应用中获得的经验。因此，它提出了动态数据捕获方法，为科学规划过程提供信息和支持。