CC BY
2
Austrian Journal of Humanities and Social Sciences 9—10 (2021) - Economics and management - ISSN 2310-5593 (Print) / ISSN 2519-1209 (Online) -
Economics and management Экономика и управление
UDK 711.7-163 DOI: 10.29013/AJH-21-7.8-50-56
S. A. KHASIKOV 1
1 Shanghai Jiao Tong University, Shanghai, China
SHARING BICYCLES - A BLESSING OR A CURSE? CASE STUDY OF DATA-DRIVEN TOOL IMPLEMENTATION DESIGNED BY URBICA FOR VELOBIKE
Abstract
Objective: the purpose of this work is to introduce and conduct a study of the tool used to increase the efficiency of sharing bicycles.
Methods: the method of cases, case study, method of specific situational analysis, method of conducting a detailed analysis of a particular situation, which is used to achieve certain research goals.
Results: this paper considers application of Big Data analysis outcomes in the field of bike sharing for improvement of VELOBIKE sharing service. VELOBIKE is a bicycle rental service with dock stations, which is operated by Moscow city authorities. URBIKA team designed and implemented data-driven tool. Compete analysis of present situation and data collected during previous seasons revealed certain patterns and allowed to visualize them. Two sided tool, which appeared to be an application with user friendly interface, was designed on the basement of these conclusions and allowed to increase amount of users trips and quality of the service.
Scientific novelty: an innovative approach of the efficiency enhancement of bike sharing service with dock stations is studied for the first time based on the methodology presented in the article.
Practical significance: the main provisions and conclusions of the article can be used in scientific and management activities of cities government when considering issues of increasing the capacity and efficiency of bicycle sharing services.
Keywords: sharing bicycle with dock station, distribution of fares, types of trips on rental bicycle.
For citation: S. A. Khasikov. Sharing bicycles - a blessing or a curse? Case study of data-driven tool implementation designed by Urbica for velobike // Austrian Journal of Humanities and Social Sciences, 2021, № 9-10. - P. 50-56. D OI: https://doi.org/10.29013/AJH-21-9.10-50-56
1. Introduction
The twenty-first century is the age of sharing culture. Residents of global megacities are ready to share many things - from umbrellas, tools and various means of transportation, to working places and even houses. Topic of this case study may seem especially surprising from this perspective - what could be more convenient
and democratic than modern comfortable bike, available anywhere at any time? It seems that this is a rhetorical question, which requires no explanations, but on practice it appeared to be impossible to ignore the context of the industrial economy transition to the information one. The idealistic idea of a bright bike as a symbol of a fair city accessible to everyone, regardless
Austrian Journal of Humanities and Social Sciences 9 - 10 (2021) ISSN 2310-5593 (Print) / ISSN 2519-1209 (Online) -
of social status and income, has broken down due to aggressive rivalry between startups in conditions of ex-cessed venture capital.
It can be argued that society was not ready for such a model of consumption, and freestanding city bicycle just outran its time. One can think that release of this product on the city streets without legislative regulatory framework and well prepared infrastructure was a premature mistake. But in this situation it is worth paying attention to the previous generation of sharing bikes, which were undeservingly forgotten. Bicycles with a dock stations have a huge undiscovered potential, which is detected due to analysis of the big data available in our technological time. This study is a perfect example of how, with the help of big data analysis and machine learning applications, Russian developers have been able to significantly increase the efficiency of "last mile" commuting, and thus make the city much more convenient for its residents. This Moscow case can become a reference example for city managers not only in China, but also all around globalizing world.
2. Stationary VS Dockless Bicycles
Prerequisites for the growth of bicycle sharing
In the 70 s, China was a "Bicycle Kingdom", but in the late 90 s, the Chinese authorities struggled to "remove" bicycles to provide the development of the automotive industry. China became so successful in setting this trend, that in 2007 it had to think again how to return the bicycle on the streets to relieve cities from the crazy car traffic [1]. The first step in this process appeared to be the classic bicycle rental with stations, but city residents
found this approach uncomfortable, which led to relatively low speed of development of such systems.
OFO and MOBIKE, that offered dockless-rent, were founded in 2014/2015, and real boom of sharing bikes has begun. Citizens appreciated cheap price of this kind of transportation, and the authorities enjoyed the opportunity to improve the climate and environment conditions in the overcrowded capital cities. Sharing cycling started to grow at an explosive pace: by the end of 2017 there were 77 sharing cycling companies in China, and the total number of available bicycles reached enormous 23 million [2].
Problems of dockless bikes
It is easy to guess that critical oversaturation of the market followed initial ultra-success. Bikes have become the scourge of Chinese cities: some streets transformed into obstacle courses because of abandoned bikes. City authorities decided to remove cycles out of the city, but the problem remained.
For most companies, oversaturation initially seemed to be a normal transition in the industry. Bicycle sharing in China has followed simple strategy: first - to attract investments to lower prices for final users; second - start to dominate the market and to raise prices for a refund [3]. However, according to financial analysts, it is impossible to reach a payback with usage of such model [4].
At the same time, investors promoted the opinion that the product ofbicycle sharing is not a lease itself, but users' data, which will be converted into targeted advertising in the long term. In the struggle for customers, many companies were dumping prices and even offering free trips, which led to the closure of most startups due to bankruptcy.
Picture 1. Chaos vs order within different models of bicycle sharing What's next? a result, by 2018, only those who opened it - OFO and
Since the end of 2017, the authorities decided to MOBIKE - remained on the market, but OFO had to sig-regulate bike sharing with restrictions and fines [5]. As nificantly reduce its turnover and it is hard to tell whether
Austrian Journal of Humanities and Social Sciences 9-10 (2021) ' - ISSN 2310-5593 (Print) / ISSN 2519-1209 (Online)
company is still alive. Meanwhile, the Chinese government is returning to the old scheme - adding more docking stations for rental bicycles.
3. Methodology of the Case Study
Case study research excels at bringing us to an understanding of a complex issue or object and can extend experience or add strength to what is already known through previous research. Case studies emphasize detailed contextual analysis of a limited number of events or conditions and their relationships. Researchers have used the case study research method for many years across a variety of disciplines. Social scientists, in particular, have made wide use of this qualitative research method to examine contemporary real-life situations and provide the basis for the application of ideas and extension of methods. Researcher Robert K. Yin defines the case study research method as an empirical inquiry that investigates a contemporary phenomenon within its real-life context [6-8]; when the boundaries between phenomenon and context are not clearly evident; and in which multiple sources of evidence are used.
Critics of the case study method believe that the study of a small number of cases can offer no grounds for establishing reliability or generality of findings [9]. Others feel that the intense exposure to study of the case biases the findings [10]. Some dismiss case study research as useful only as an exploratory tool [11]. Yet researchers continue to use the case study research method with success in carefully planned and crafted studies of real-life situations, issues, and problems [12]. Reports on case studies from many disciplines are widely available in the literature.
Many well-known case study researchers such as Robert Stake, Helen Simons and Robert Yin have written about case study research and suggested techniques for organizing and conducting the research successfully. This introduction to case study research draws upon their work and proposes six steps that should be used:
• Determine and define the research questions;
• Select the cases and determine data gathering and analysis techniques;
• Preparation for data collection;
• Collection of the data in the field;
• Evaluation and analyses of the data;
• Preparation of the report.
Case studies are complex because they generally involve multiple sources of data, may include multiple cases within a study, and produce large amounts of data for analysis. Researchers from many disciplines use the case study
method to build upon theory, to produce new theory, to dispute or challenge theory, to explain a situation, to provide a basis to apply solutions to situations, to explore or to describe an object or phenomenon. The advantages of the case study method are its applicability to real-life, contemporary, human situations and its public accessibility through written reports [13-14]. Case study results relate directly to the common reader's everyday experience and facilitate an understanding of complex real-life situations. 4. VELOBIKE Case Study What is VELOBIKE and who are URBICA? "Velobike" is a bicycle sharing system with doc stations run by the city of Moscow, Russia. Moscow is the largest urban agglomeration in Russia and Europe with a permanent population of about 17 million people. As most capital cities of developing countries, Moscow is also suffering from overpressure on existing transportation system. "Velobike" is one of the tools, specially developed for mitigation of car traffic in city center and providing commuters alternative ecological public transport. Main costs of this service are covered by institutional investors while operating costs are covered by the city of Moscow. The system is successfully working since its opening in 2013. In 2018 this bicycle sharing service had 424736 registered users and 4.25 million performed journeys during summer season [15].
According to official web site, URBICA is Moscow based team of developers, cartographers, designers and data analysts, who are focusing on human experience in the cities [16]. URBICA is successfully working in the fields of information design, user interface design, service design, architecture, environmental design, data analysis and ethnographic research.
URBICA team was given the task to improve operation ofVelobike and solve one of most critical problem, that can significantly deteriorate users experience from the whole service - as it grew more popular in Moscow, so did the frequency of situations when users ran into the problem of zero available bikes or zero empty docks. Such phenomena of empty or full stations is called "dockblocking" [17-18]. Analyses of present situation Work on the project of optimization of the city bicycle rental network started from the statistics analysis of users rides. The data, provided by the Department of transport of Moscow city, contained information on the duration of the trip and what fares were used for each one of them. Analyses allowed to identify following patterns of fare types: "Day" fare is used for long rides on the em-
Economics and management - ^E.M|ER Austrian Journal of Humanities and Social Sciences 9—10 (2021) - ISSN 2310-5593 (Print) / ISSN 2519-1209 (Online) -
bankments and across the historical touristic downtown (34% of the total number of trips); "Season" fare is more often used for short and regular trips between districts (55% of all trips); "Month" fare appeared to be not so popular - only 11% of all trips (Picture 2).
One of the parameters that also needs to be considered is distribution of fares for each station on different
days of the week. The most popular fare on weekends is "Day", but on weekdays it appeared to be "Season". Obvious dependence of the bike rental on weather conditions is also confirmed - cyclists with the "Day" fare are much less likely to ride a bike in rainy weather and at low temperatures, but users of "Season" and "Month" fares are almost not bothered by bad weather (Picture 3).
Picture 2. Percentage of trips within district, distribution of fare types and duration of rides Revealed fare patterns and its distribution, trips dura-
tion and priority of fare usage according to the day of the week allowed to define three main types of rides:
1. "Ride for fun" (24% of total trips) - cycling with a return to the same station. This type of activity in residential areas is popular not only on weekends, people enjoy short time bicycle rides near their houses even after work.
2. "Trip between metro stations" (14% of total trips) - a ride started and completed near two metro stations. This trip is not always a replacement of the metro; it can be a ride around the city. The average trip time on weekdays - 35 minutes, and on weekends - 48 minutes.
-MOn5V- 'ant -S*»SOr- ■ toTTirnlure, C* • rimydav
Picture 3. Distribution of trips at different fares for each day
Austrian Journal of Humanities and Social Sciences 9 - 10 (2021)
Economics and management
ISSN 2310-5593 (Print) / ISSN 2519-1209 (Online)
3. "Trip from metro to address" (20% oftotal trips) -rides, that usually begin near metro stations, and end at the bike stations, which are not. In fact, this type of travel is so called "last mile", replacing the bus or long walk. The average duration of such ride on weekdays is 26 minutes.
The next stage of a more detailed study, for application of a long-term strategy of cycling development in conjunction with the entire transport system of Moscow, was the analysis of the docking stations. Data on the operation of the service allowed to cluster stations into four groups: stations getting empty in the morning and getting refilled by the evening; stations collecting bicycles in the morning and getting empty again in the evening; stations remaining empty throughout the whole day; station collecting bicycles throughout the day. These groups can be combined into two pairs that follow consequential patterns (Picture 4).
First pair of groups can be explained by commuting during weekdays - dock stations located inside residential districts get empty in the morning; at the same moment, stations near subway collect these bicycles. On the contrary, stations located in the business districts, get empty close to the subway, but amass bikes near office buildings every morning. The opposite situation can be observed in this groups in the evening.
Second pair can be combined together due to the glance at the topography map. Terrain makes some stations constantly loose bicycles, and some constantly accumulate them. Station that is situated on the famous Vorobyov Hills viewport is in high demand among users who are willing to enjoy their ride down to the embankment of the Moscow river. On contrary, there are almost no cyclists who are willing to return a bicycle to a dock station here.
Picture 4. Self-rebalancing stations. Stations continuously accumulating and expending bikes
5. Solution
How does URBICA solves problem of "dockblocking"?
After complete analysis of the bike rental service statistics, URBICA team created a demand forecasting algorithm to help track and dispatch bikes among stations, so that they are always sufficiently supplied. It is a two-sided tool with a convenient user interface that predicts station demand and consists of a dispatcher to track bike availability and a Telegram Bot that sends automatic notifications to drivers and issues rebalancing tasks (Exhibit 5). In practical terms, operation teams can "rebalance" the system - shift bikes from full stations to empty ones. However, employing a dispatcher that guides the transfer of bikes with the help of special trucks strategically navigated to pick-up and drop-off points around the city amounts to considerable expenditure on maintenance.
How is this process organized?
A supervisor surveys a map that displays bike availability at every station and decides which truck to direct to which station and the amount of bikes to be redistributed. Trucks have regular routes at night in order to
prepare high demand stations for the morning rush, but when demand increases above normal levels or follows an abnormal pattern, dispatchers and fleet technicians are not able to provide stations with the needed amount of bikes. Without reliable prediction tools, the operations team can only react to the current situation based on their own experience.
Forecasting demand
The developed prediction algorithm takes into account several different input parameters, including the historical availability of bikes at every station per hour and the distribution of ridership. This is sufficient data to inform fleet techs about stations in need of bikes and stations likely to be full in order to balance the system.
According to URBICA report, the system was tested during summer season of 2018 and, in the process, it turned out that many of the operations normally performed by a manager can be easily automated. With the help ofmachine learning, it is possible to free up time for dispatchers and optimize logistics which may help reduce the quantity of automobiles and reduce the cost of rebalancing.
Austrian Journal of Humanities and Social Sciences 9 - 10 (2021) ISSN 2310-5593 (Print) / ISSN 2519-1209 (Online) -
13:17
# 0035
• » ' P - <
• *■ * • • .
■ • ». •
ft Ky im-.;eiih MÚL1 i'y ¿ ï L.*>. L
BarpyaieHHöeTb
* • %
♦ ■
*
¿0033
if 0034
fl 003S
* ■ m
«A m * •
* !•
HMCÜKW
• .
Pi 0037
—. , ,: Il i- ll#ílft№l
*
• •
''i * • • • *
•••
• ». .
i* ti i* ti ti
KuuMenupM:^
Picture 5. General user interface for supervisors
At the end of 2014 rental scored a little more than 100 thousand trips. Due to URBICA tool, users made 2.4 million trips at the 380 stations at the 2018, and it is 50% more than previous season.
6. Conclusion
The example of the "Velobike" case demonstrates an interesting paradox: an in-depth meaningful analysis and return to operation of the previous generation of rental bicycles, which seemed to be technically outdated and did not meet the needs of modern citizens, allowed them to take a worthy place in the future - bicycles with dock stations took a step back, in order to make two steps forward later. Unfortunately, understanding that the consumer society needs to get prepared for sharing culture is constantly increasing in 2019. The variety of multicolored bicycles, which seemed to be a panacea in the field of ecological transportation, threatens to become another man-made apotheosis of human greed - the scale of the so-called "temporary bicycle storage" parking lots are capturing imagination. It is impossible not to notice the contradiction between their widespread appearance and the generally accepted declaration of value and respectful exploitation of natural resources. A tool
designed to make the city more convenient and eco-fnend-ly, enthusiastically accepted by both ordinary citizens and city management, was too rashly and prematurely placed on the banner of the sharing culture, discrediting and causing significant damage to the idea itself.
Of course, it would be too foolish to deny that freestanding city bike is the future. But at this stage, in addition to the lack of a culture of such a product usage, the infrastructure, as well as the legislative framework, are also vitally needed. The formation of the abovemen-tioned conditions will take time, and exactly on this transit period flourishing of bicycles with docking stations should come. The "Velobike" case proves that the use of a modern analytical approach based on big data and machine learning can significantly improve the efficiency of the service, and as a result, make the city more comfortable for its residents. Bicycles tied to stations can also be convenient and affordable. Proper planning of docking stations location and well-planned flexible work of operational teams are able to bring the service to a new level that fully meets the needs of residents of megacities of the twenty-first century.
References
1. Poddubchak Y. Veloshering v Kitae: bol'she ne znachit luchshe (Bikesharing in China: more does not always mean good), 2019. (In Russian). Available from: URL: https://truesharing.ru/tp/16817
Austrian Journal of Humanities and Social Sciences 9—10 (2021) - PREM|ER Economics and management
- ISSN 2310-5593 (Print) / ISSN 2519-1209 (Online) -
2. Deljukin E. Vzljot i padenie kitajskogo rynka arendy velosipedov: istorija Ofo i Mobike (Up's and down's of bikesharing market: the story of Ofo and Mobike), 2019. (In Russian) Available from: URL: https://vc.ru/ transport/57326-vzlet-i-padenie-kitayskogo-rynka-arendy-velosipedov-istoriya-ofo-i-mobike
3. Mobike Global. Recycling, refunding and redesigning: unveiling mobike's new growth initiatives, Mobike, 2018. Available from: URL: https://www.mobike.com/global/blog/post/growth-initiatives
4. Romashkevich A. Veloprokat bez stancij : vesti s polej (Bikesharing without dock stations: latest news). Velonation, 2018. (In Russian). Available from: URL: https://velonation.bike/2018/03/05/docklessbike
5. Kunzing R. To build the cities of the future, we must get out of our cars, National Geographic, 2019. Available from: URL: https://www.nationalgeographic.com/magazine/2019/04/to-build-cities-of-the-future-stop-driv-ing-cars
6. Yin R. K. Case Study Research: Designs and Methods, SAGE: - London, Thousand Oaks, New Deli, 1989.
7. Yin R. K. The Case Study Anthology, Thousand Oaks, CA: Sage, 2004.
8. Yin R. K. Case Study Research: Designs and Methods, Thousand Oaks, CA: Sage, 2009.
9. Eisenhardt K. M. Building theories from case study research, Academy of Management Review,- 14(4). 1989.
10. Ellet W. The case study handbook: How to read, discuss and write persuasively about cases, - Boston (MA): Harvard Business School Press. 1989.
11. Easton G. Critical Realism in Case Study Research, Industrial Marketing Management,- 39, 2010.
12. Naumes W. The Art & Craft ofCase Writing by William Naumes and MargaretJ. Naumes. M. E. Sharpe, Inc. 2006.
13. Stake R. E. The Art of Case Study Research, Thousand Oaks, CA: Sage, 1995.
14. Stake R. E. Qualitative Case Studies / Denzin N., Lincoln Y.- 3rd edition, Thousand Oaks, CA: Sage, 2005.
15. Dolgova A. Sistema besstancionnogo veloprokata mobike vot-vot pojavitsja v Rossii. Chto o nej nuzhno znat'? (The Mobike bike rental system is about to appear in Russia. What do you need to know about it?), University "Strelka", 2018. (In Russian). Available from: URL: https://strelkamag.com/ru/article/mobike
16. URBICA, Improving Moscow's bike share, - Urbica, 2018. Available from: https://medium.com/@Urbica.co/ improving-moscows-bike-share-58007a932d5
17. URBICA, Velobike, - Urbica, 2017. Available from: URL: https://v3.urbica.co/velobike
18. URBICA, Velosipedi v gorode (Bicycles in the city), - Urbica, 2015. (In Russian). Available from: URL: https://medium.com/russian/BeAOCHneAM-B-ropoAe-230eb284b244
Information about the author
Sergei Khasikov, Shanghai Jiao Tong University, Shanghai
Address: Shanghai, Minhang District, Dongchuan Street, 800, China
E-mail: khasikovsergei@yandex.ru; Tel.: +86 21 54-74-00-00
ORCID: 0000-0002-3993-6186
