During my educational travels, I've experienced public transport systems in various countries. In developed nations like Germany, public transport is characterized by the comfort of fixed schedules and reliable routes. However, it often feels impersonal. In contrast, the unorganized transport in India, despite its chaos, offers a unique sense of belonging. Engaging in the comical dance of bargaining over trip costs, asking drivers to stop a few extra meters before or after the 'designated' stops for better last-mile connectivity, or witnessing multiple transport providers competing to serve you, brings a personal touch to the experience. And despite this chaos - or maybe because of it - the system works.

Unorganized, informal transport is ubiquitous not just in India but across most countries in Asia, Africa, and South America. It encompasses a broad array of motorbike or minibus services that act as the primary form of motorized mobility for 80% of the world's population residing in the Global South. These are dynamic systems operated by private individuals seeking a livelihood. They are characterized by drivers who adjust routes, frequency, and operating conditions on the fly. The route and operating conditions of these services grow organically through interaction among competing service providers, users, and changing city structures. The service provider aims to maximize profit, often leading to bad vehicle conditions and reckless driving.

The private nature of these services has limited the research available on these systems because data is scarce and operations are decided on an individual level without major central planning and record keeping. Unlike centrally planned systems, informal transport may self-organize into different operational settings. However, it remains unclear how drivers make decisions, interact with users, determine operating conditions, stay profitable while competing with subsidized, more organized alternatives in many regions, and how the decentralized structure of these organizations impacts the routes and trips of these services.

Given the large number of interacting parameters, these questions have been challenging to answer due to the complexity of modeling these systems. Together with the perceived chaotic conditions, the lack of knowledge has led to a negative image of informal transport, prompting a shift towards more formal alternatives as countries in the Global South develop. However, recent technological advancements and increased data collection efforts are providing an increasing set of data on the route structure of informal transport. These new data have enabled us to systematically analyze the route structure to understand the underlying properties of these systems.

In our research, we found that the routes of informal transport are far from random; they self-organize into efficient structures. We analyzed over 7000 routes in 36 cities and measured the detours along each route compared to direct trips, comparing routes from formal and informal services across the globe. To quantify the structural aspects of routes, we define their directness in terms of the total detour compared to direct (taxi-like) routes, the social fairness of how detours are distributed along the routes, and the interconnectivity of the route network. Typical bus routes are relatively straight in the middle of the route and have more detours towards their ends. Surprisingly, the majority of informal transport routes in the Global South behave differently. The routes are overall more direct with fewer and more fairly distributed detours. Overall, informal transport routes thus seem to often outperform the public transport routes of the Global North in structural efficiency while remaining profitable, highlighting the potential of self-organization for mobility services. 

Our analysis shows that in many cities, the apparent chaos of informal transport is actually a symphony of self-organization. We should try to learn from the successful parts of such self-organization processes to design more efficient formal transport routes and to dynamically adapt services to changing conditions in changing urban environments. We therefore need to understand how users and service providers interact, how the route structures emerge from these complex interactions, and where self-organization fails to yield efficient routes.

Our work may serve as a novel starting point for studying the structural properties of informal transport routes. It already demonstrates that blindly replacing informal transport with more formal alternatives without prior data-based analysis may not necessarily provide better public transport. Further analyses - and more data, particularly on the temporal aspects of transport - are still needed to understand aspects of informal transport such as frequency, waiting time, and safety conditions. Informal transport can help add the personal touch that formal transport requires, promoting shared mobility and helping make our cities across the globe more livable in the process.