Optimal rain gauge network to reduce rainfall impacts on urban mobility – a spatial sensitivity analysis
| DOI | https://doi.org/10.1108/IMDS-03-2022-0145 |
| Published date | 18 July 2022 |
| Date | 18 July 2022 |
| Pages | 2261-2280 |
| Subject Matter | Information & knowledge management,Information systems,Data management systems,Knowledge management,Knowledge sharing,Management science & operations,Supply chain management,Supply chain information systems,Logistics,Quality management/systems |
| Author | Felipe de Oliveira Simoyama,Lívia Rodrigues Tomás,Felipe Matheus Pinto,Luiz Leduino Salles-Neto,Leonardo Bacelar Lima Santos |
Optimal rain gauge network to
reduce rainfall impacts on urban
mobility –a spatial
sensitivity analysis
Felipe de Oliveira Simoyama
Instituto de Ci^
encia e Tecnologia (ICT), UNIFESP, Osasco, Brazil
L
ıvia Rodrigues Tom
as
CEMADEN, S~
ao Jos
e dos Campos, Brazil
Felipe Matheus Pinto
UDESC, Lages, Brazil
Luiz Leduino Salles-Neto
Instituto de Ci^
encia e Tecnologia (ICT), UNIFESP, Osasco, Brazil, and
Leonardo Bacelar Lima Santos
CEMADEN, S~
ao Jos
e dos Campos, Brazil
Abstract
Purpose –A sustainable transportation system should represent a win-win situation: minimizing transport’s
impacton the environmentand reducingnatural disasters’effectson tra nsportation. A well-distributed set of rain
gaugesis crucial formonitoring servicesin smart cities.However, thoseservices shouldconsider theuncertainties
aboutthe registersof rainfallimpacts. In thispaper, the authorspresenta case study of optimalrain gaugelocation
based on an actual database of rainfall events with impacts on urban mobility in the city of Sao Paulo (Brazil).
Design/methodology/approach –This paper presents a maximal covering location formulation and
proposes a robustness analysis considering spatial location perturbations.
Findings –In this case study, the robustness of the objective function is above 99.99%. The robustness for the
number of covered demand points is 88.93%, and the frequency associated with every candidate is between
11.71% and 69.49%.
Originality/value –Incorporating spatial uncertainties on coverage problems is essential to provide
stakeholders more realistic supporting tools and to draw different possible scenarios.
Keywords Rain gauge network, Maximal covering, Smart cities, Optimization
Paper type Research paper
Introduction
The occurrence of extreme events around the world has significantly increased in recent
years. Climate change is critical in this context: extreme meteorological events will be even
Rain gauge
network –
sensitivity
analysis
2261
The authors thank Claudia Linhares and Demerval Goncalves (Cemaden), researchers from CGE
(Emergency Management Center of the city of Sao Paulo), and from Climatempo for the valuable
discussions.
The authors would like to thank the S~
ao Paulo Research Foundation (FAPESP-Brazil) [grant
numbers 2021/03269-7, 2016/01860-1], the National Council for Scientific and Technological
Development (CNPq-Brazil) [grant numbers 405702/2021-3, 304528/2021-8] and the Coordination of
Improvement of Higher Education Personnel (CAPES-Brazil) for the financial support.
Funding: This study was financed in part by the Coordination of Improvement of Higher Education
Personnel (CAPES-Brazil) - Finance Code 001, the S~
ao Paulo Research Foundation (FAPESP-Brazil)
[grant numbers 2021/03269–7, 2016/01860–1] and the National Council for Scientific and Technological
Development (CNPq-Brazil) [grant numbers 405702/2021–3, 304528/2021–8].
The current issue and full text archive of this journal is available on Emerald Insight at:
https://www.emerald.com/insight/0263-5577.htm
Received 9 March 2022
Revised 2 May 2022
26 May 2022
Accepted 29 June 2022
Industrial Management & Data
Systems
Vol. 122 No. 10, 2022
pp. 2261-2280
© Emerald Publishing Limited
0263-5577
DOI 10.1108/IMDS-03-2022-0145
more frequent and severe (Pachauri et al., 2014;UNISDR, 2015). Additionally, vulnerability to
those extreme meteorological events is even more severe in developing countries (Nur and
Shrestha, 2017), and impacts on transportation infrastructure and urban mobility are high
and increasing (Santos et al., 2017;Carvalho et al., 2018;Lamosa et al., 2021).
In Brazil, for example, there are approximately 40 thousand areas at risk of events
triggered by rainfall, such as floods and landslides. However, only 3% of these areas are
monitored by the responsible bodies (Alert a contra inundaç~
oes, 2018). Sao Paulo (Brazil) is
the largest city of the country and of both southern and western hemispheres, with a
population of more than 12 million inhabitants (F. I. B. de Geografia e Estat
ıstica, 2020), and
also the largest gross domestic product (GDP) in Latin America (BRL 699.28 billion) (F. I. B. de
Geografia e Estat
ıstica, 2019). The city of Sao Paulo suffers from its unplanned urbanization
process, which resulted in an increasing population density and a vast transformation of
green areas into urbanized areas (Multini et al., 2016). Several factors contribute to the high
number of disasters related to extreme rainfall events in Brazil, such as the spatial
distribution of the population across the territory, and natural susceptibility of the terrain (de
Assis Dias et al., 2018).
One of the biggest concerns of the population in several big cities, such as Sao Paulo, is
urban mobility. There are 8,733,790 vehicles registered in the city, which corresponds to
8.17% of the total number of vehicles in Brazil. Rio de Janeiro has the second largest fleet of
registered vehicles in Brazil, with 2,935,084, i.e. one-third of the registered vehicles of Sao
Paulo (M. da In fraestrutura do Brasil, 2020). On average, a Sao Paulo inhabitant spends
2 hours and 25 minutes with their daily commutes (I. Intelig^
encia, 2019). Vale (2020) says that
opportunity costs arising from urban immobility can reach BRL 7.3 billion per year in
Sao Paulo.
According to data from the Companhia de Engenharia de Tr
afego (CET) (CET, 2020), the
Traffic Engineering Company of S~
ao Paulo, in 2019, the city averaged 77 and 85 kilometers of
traffic in the morning and afternoon peak hours, respectively. Massive traffic events can be
the consequence of several factors: the increased number of vehicles, unplanned urbanization,
accidents on the roads, heavy precipitations and flooding. Heavy precipitations usually result
in slower traffic and more dangerous traffic conditions, since it causes bad visibility and
changes road friction (Litzinger et al., 2012).
Floods are an increasing concern in big cities and pose several challenges to urban
livability (Miguez and Ver
ol, 2017) and businesses (Tse et al., 2016). Estimates say that floods
in the city of Sao Paulo caused average losses of BRL 226 million per year between 2008 and
2012 (Teixeira and Haddad, 2008). Such estimates take into consideration the people and
firms directly affected and the impact on the national economy, since impacts in Sao Paulo
can produce broader effects to other areas of the country (Haddad and Teixeira, 2015).
Besides of the economic losses, floods caused by heavy rainfall and drainage system
issues lead to urban immobility, especially in urban areas. The resident population in flooded
areas are directly affected by floods because they might be ‘trapped’or immobile due to
disruptions in the transportation network (Ayeb-Karlsson et al., 2020). Additionally,
commuters are also exposed because they might be inadvertently at risk during severe
weather, since they are not able to cancel their regular trips voluntarily (Liu et al., 2021).
In this scenario, a well-distributed set of rain gauges on the urban space is crucial for
monitoring services in smart cities, including early warning systems, civil defenses and
traffic engineering companies. However, those services should consider the uncertainties
about historical records. Traditionally, the register of extreme rainfall events brings a
reference point only, not a detailed description of the whole area. It is possible to associate an
uncertainty component in this problem, incorporating perturbations on each event’s spatial
location. But how does t he optimal location pr oblem formulation pr opagate those
perturbations?
IMDS
122,10
2262
To continue reading
Request your trial