Login

Project

#366 Smart and equitable parks: quantifying returns on investments based on probabilistic mobility-dependent correlates of park usage using cyber-physical system technologies


Principal Investigator
Katherine Flanigan
Status
Completed
Start Date
July 1, 2021
End Date
June 30, 2022
Project Type
Research Applied
Grant Program
FAST Act - Mobility National (2016 - 2022)
Grant Cycle
2021 Mobility UTC
Visibility
Public

Abstract

A challenge for local governments is to develop and maintain parks in ways that equitably distribute benefits to health, well-being, livability, accessibility to essential services, and the economy.  The overarching goal of this proposal is to explore urban park use and correlates of use related to mobility services (i.e., their quantitative linkage to economic growth through equitable access to essential services) using cyber-physical system technologies in order to bring to light ways in which city officials and planners can quantify data-driven returns on potential investments to parks and mobility services and implement changes that will uniformly distribute these benefits.    
Description
Parks are integral to the success of any vibrant city and have long been touted as engines of economic growth that also improve public health, clean the air, manage stormwater, and enable patrons to commune with nature while enjoying a rich set of social experiences within their community [1–2].  High-quality park systems attract employers, accelerate the pace of commercial development, and generate new tax revenue.  Today, 165 parks are maintained in Pittsburgh ranging from small neighborhood parks to large greenways.  Evidence of the distribution of parks is that approximately 91% of Pittsburgh residents live within a half-mile walk of a park [3].  Unfortunately, the financial constraints of the city have challenged its ability to maintain its parks.  Despite relatively good access to green space, Pittsburgh parks are underinvested in comparison to both regional and aspirational peers [3].  Local governments also face challenges associated with urban patterns of social segregation, community disinvestment, and disproportionate access to public goods [4].  Thus, a key challenge for local governments is to develop and maintain parks and other public goods in ways that equitably distribute benefits to health, well-being, livability, accessibility to essential services, and the economy.  Here, essential services range from healthcare (e.g., hospitals, pharmacies) to fresh food distributors (e.g., grocery stores, SNAP retailers) to schools with positive student outcomes, just to name a few.

While it is well understood that there exists a correlation between parks and surrounding economic growth, there is yet to emerge a quantitative mechanism that can integrate results of public park use and quality measurements into an overall evaluation scheme suitable for practical use by planners and city officials.  Our knowledge of a broader range of urban parks——particularly small and medium-sized community parks which may be integral to the daily routines of urban residents——is especially limited [4].  This is in part due to the historical public facility management paradigm cast as a problem of achieving equity based on static spatial distribution metrics.  Existing planning and maintenance decisions aim to evenly distribute parks among populations spatially with little to no consideration of the location or measurable quality of mobility services that link populations to essential services via parks [5–11].  As a result, in areas where essential services are unevenly distributed across a community, parks and greenways often lead to a bifurcation: they either serve as barriers that can result in social polarization, or serve as enabling public facilities that can connect citizens in under-resourced areas to their wider communities and services; the polarizing or unifying nature of parks is heavily dependent on the configuration and health of surrounding mobility services.  In the context of this proposal and without loss of generality, mobility services will refer to public transportation and walkable/bikable pathways, which can be planned and developed to positively affect usage patterns.  A number of initiatives in Pittsburgh are already exploring various ways in which mobility services can be redesigned to affect how people move through communities.  For example, every summer OpenStreetsPGH [12] temporarily closes streets to automobile traffic and invites Pittsburghers to reimagine their streets as places for people and to explore new areas.  OpenStreetsPGH is inspired by the global open streets movement in which cities across the globe celebrate open streets to promote car-free transportation, healthy outdoor activity, and community engagement using city streets and park facilities.

The overarching goal of this proposal is to explore urban park use and correlates of use related to mobility services (i.e., their quantitative linkage to economic growth through equitable access to essential services) using cyber-physical system (CPS) technologies in order to bring to light ways in which city officials and planners can quantify data-driven returns on potential investments to parks and mobility services and implement changes that will uniformly distribute these benefits.  This goal acknowledges that the effectiveness of parks and greenways is highly dependent on their ability to connect residents to essential services and their surrounding community, and respects that the reliability of these key transportation linkages should not be limited to just severe disasters.  It should also consider everyday disturbances (e.g., delays in public transportation and bad weather) that can be measured using emerging wireless sensing technologies to guarantee an acceptable level of service, even when the capacity of certain roadways and pathways is degraded by various physical and operational problems.

The research team will approach this problem through the development of a novel “sensing for decisions” framework, which refers to data-driven decision processes based on sensed data that have quantifiable returns on investment.  The proposed framework will follow a network theoretic approach in which a community’s network topology and transportation properties are represented as interconnected links and nodes with stochastic attributes that can be characterized directly from in situ sensing data: discretized subsets of a community’s population, parks (weighted as a function of their health and effectiveness), essential services (weighted as a function of the type of service provided), and connecting transportation and mobility services (weighted as a function of their health and effectiveness).  The system performance is evaluated using a data-driven probabilistic network reliability analysis.  The output of the network reliability analysis is the system-level probability of failure with respect to a given failure limit state (described subsequently), where degradation or failure of transportation linkages and the reduction of nodal weights increases the network’s probability of failure.  From the perspective of city planning officials, the objective is to strategically invest in the nodes and linkages that they have control over——meaning parks and connecting mobility services——in order to minimize the network’s probability of failure.  That is, they effectively “actuate” controllable parts of the network through maintenance, rehabilitation, and capital projects (such as the development of optimally placed arterial greenways) to minimize the system-level probability of failure given cost constraints.  Notions of equity are deeply rooted within the proposed network reliability analysis based on how the failure limit state is defined: a community’s system-wide failure is defined as any subset of a community’s population not being able to access and/or benefit sufficiently (if at all) from its surrounding community’s resources.  This ensures that targeted investments are made to reduce a community’s network probability of failure subject to the constraint that all subpopulations must remain above an acceptable accessibility threshold.  For the purpose of making optimal urban planning and investment decisions, through this framework we will be able to simulate the effects of specific park and mobility service maintenance, rehabilitation, and capital projects on the probabilistic health and weights of network nodes and linkages, as well as the connectivity of the network.  Given changes proposed by city officials and the public (such as through the open-source software platform Kurb [13]), we will be able to explicitly quantify returns on investments through corresponding shifts in the magnitude of the network probability of failure.

An additional novel contribution of this work is that the probabilistic health and usability of walkable/bikable transportation corridors that are modeled as linkages in the aforementioned network reliability problem will be directly measured using low-power wireless sensing nodes already developed by the PI [14].  Each sensing node, called Urbano (from the Latin urbanus, “of or belonging to a city”, derived from urbs, “city”), is an Internet of Things- (IOT) based technology that supports interoperability among diverse arrays of heterogeneous IoT devices, preserves privacy and trust among citizens, supports cloud-based analytics, and has a user-friendly design.  For example, discrete pedestrian counts (measured from passive infrared sensors) collected from distributed sensing nodes will be used to indirectly infer the health of continuous pathways.  Because Urbano nodes support low-power and low-cost sensing and use cellular communication to free nodes from fixed power sources (relying instead on small solar panels for solar harvesting), they can be deployed in under-resourced areas, enabling decision makers to make data-driven investments in neighborhoods that have historically been underserved.  This addresses the urgent need for quantitative approaches to integrated park and transportation design and management in under-resourced areas to ensure that investments made in parks using limited tax dollars have maximum effect.  In situ data measured using Urbano nodes will be transmitted to a cloud-based server and will provide a wealth of information that will be used to characterize the probabilistic health of key transportation linkages based on, for example, daily environmental (i.e., weather) and operational conditions.  The impact of deterioration on mobility services manifests in the network’s probability of failure.  Additionally, Urbano nodes will be configured to measure air quality (NO2, SO2, O3, and PM2.5) and noise intensity at parks within the network, which will directly influence their nodal weights within the network model.

To provide a concrete example illustrating the importance of data-driven network analysis based on in situ measurements, during the late summer and fall of 2018, the PI used a network of Urbano sensing nodes to monitor pedestrian traffic on a section of a vital pathway connecting an under-resourced community to the Detroit Riverfront (Detroit, MI) [14].  Pedestrian use data was collected in real time and several use trends and correlations were observed including correlations between weather and trail usage as well as the numbers of pedestrians at various times of the day.  A series of unexplained periods of drastically reduced trail usage were also observed.  After visiting and investigating the location it was determined that these periods of little to no pedestrian traffic were correlated to unexpected (i.e., not anticipated by the urban planners) severe flooding along the trail due to the storm sewer inlet infrastructure not functioning properly; this critical pathway was rendered unusable for several days following any rain event and the collected data prompted rehabilitation efforts.

While the project described in this proposal is applicable to any park in the city (or state for that matter), the team has confirmed support from several partners to carry out a full-scale proof of concept.  The project team will partner with (see attached letters of support):

1. the City of Pittsburgh Department of Public Works
2. the City of Pittsburgh Department of Parks and Recreation
3. the City of Pittsburgh Department of Mobility and Infrastructure
4. and the Pittsburgh Parks Conservancy,

who will contribute to the project by providing guidance and resource support (i.e., permission to install Urbano sensing nodes on light poles) for the deployment of the project.  With the support of Metro21 functioning as a community-research liaison (see attached letter of support), the project team will work with the City of Pittsburgh and Pittsburgh Parks Conservancy to select a small to medium-size park and surrounding community to function as a testbed.  Through our partners, we have been given full access to install a distributed network of Urbano sensing nodes within the park and around the surrounding community.  The research team is uniquely positioned to implement the proposed research goal within a Pittsburgh community due to the PI’s previous work in data-driven system reliability methods and urban sensing [14–16].  The PI’s theoretical work has been augmented with deployments and experimental work that have reached full scale.  In addition to leading the development and deployment of a long-term structural health monitoring program on the Harahan Bridge near Memphis, TN, the PI has worked with community and government stakeholders to: 1) use GPS-enabled Urbano nodes installed on the roofs of food trucks to assess compliance with permit rules and curbside management; 2) use Urbano nodes to observe the utilization of public spaces; 3) use Urbano in combination with a structured curriculum (called Sensors in a Shoebox) to empower Detroit communities to observe their neighborhoods and analyze sensor data to address fundamental questions about the complex urban issues they face.

References:

[1] Neema MN and Ohgai A. Multi-objective location modeling of urban parks and open spaces: continuous optimization. Computers, Environment and Urban Systems 2010; 34(5): 359–376.
[2] Pittsburgh Parks Conservancy. Benefits of Pittsburgh’s Parks System, https://pittsburghparks.org/wp-content/uploads/2020/10/PPC_RPP_Benefits_Case_0911-4.pdf (2019, accessed 10 December 2020).
[3] The Trust for Public Land. Pittsburgh, PA ParkScore Ranking, https://www.tpl.org/city/pittsburgh-pennsylvania (2019, accessed 10 December 2020).
[4] Hamstead ZA, Fisher D, Ilieva RT et al. Geolocated social media as a rapid indicator of park visitation and equitable park access. Computers, Environment and Urban Systems 2018; 72: 38–50.
[5] Jones K and Kirby A. Provision and wellbeing: an agenda for public resources research. Environment and Planning A 1982; 14(4): 297–310.
[6] Kirby A, Knox P and Pinch S. Developments in public provision and urban politics: an overview and agenda. Area 1983; 15(4): 295–300.
[7] Pinch S. Inequality in pre-school provision: a geographical perspective. In: Kirby A, Knox P and Pinch S (eds) Public service provision and urban development. London: Croom Helm, 1984, pp. 231–282.
[8] Smith DM. Geography and Social Justice. Oxford: Blackwell, 1994.
[9] Hay AM. Concepts of equity, fairness and justice in geographical studies. Transactions of the Institute of British Geographer 1995; 20: 500–508.
[10] Talen E and Anselin L. Assessing spatial equity: an evaluation of measures of accessibility to public playgrounds. Environment and Planning A 1998; 30: 595–613.
[11] Ogryczak W. Inequality measures and equitable approach to  location  problems. European Journal  of  Operational Research 2000; 122: 374–391.
[12] BikePGH. OpenStreetsPGH, https://openstreetspgh.org/ (2020, accessed 10 December 2020).
[13] Headstorm. Kurb, https://kurb.io/ (2020, accessed 10 December 2020).
[14] Flanigan KA, Handley J, Lynch JP, et al. Open urban sensing architecture for community-governed smart and connected communities. Submitted to Sustainable Cities and Society.
[15] Flanigan KA, Lynch JP and Ettouney M. Probabilistic fatigue assessment of monitored railroad bridge components using long-term response data in a reliability framework. Journal of Structural Health Monitoring 2020; 19(6): 2122-2142.
[16] Flanigan KA, Lynch JP and Ettouney M. Quantitatively linking long-term monitoring data to condition ratings through a reliability-based framework. Journal of Structural Health Monitoring 2020; doi: 10.1177/1475921720949965.
Timeline
July 2021 through September 2021

During the first quarter of the project, the project team will:

Prepare the testbed by selecting park/surrounding community (in collaboration with the City of Pittsburgh, Pittsburgh Parks Conservancy, and Metro21), building Urbano nodes in lab, rigorously testing Urbano nodes in lab, deploying Urbano nodes in testbed community, and configuring the web application for visualization and descriptive analysis.  Note: collecting data towards the beginning of the project is a high priority to ensure that the collected data can fully capture correlations to changes in weather, operational, and usage patterns over the course of the project.

October 2021 through March 2022

During the second and third quarters of the project, the project team will:

Focus on the first primary intellectual contribution, which is the formalization of the data-driven probabilistic network reliability problem that accounts for stochastic link and nodal weights.  Focus on the second intellectual contribution, which is the formalization of a lower limit state function that accounts for a community’s system-wide failure as defined as any subset of a community’s population not being able to access and/or benefit sufficiently (if at all) from its surrounding community’s resources.  This lower limit state function is integral to accounting for equity in the system performance metric.

April 2022 through June 2022

During the final quarter of the project, the project team will:

Focus on the third primary intellectual contribution, which is the application of the theoretical framework to the testbed using sensed data (which will have been collected for about 7 months at this point).  Use GIS map data from testbed to model the community’s network topology and transportation properties.  The majority of the effort in the final quarter will be spent processing pedestrian and environmental data to probabilistically characterize the linkage and nodal weights, which serve as inputs to the analytical framework.  Summarize findings in a final report and present recommendations to collaborating partners and stakeholders (i.e., City of Pittsburgh and Pittsburgh Parks Conservancy).  Demonstrate to the City how their recommendations, as well as the public’s recommendations (using, for example, the open-source Kurb platform [13] and OpenStreetsPGH [12]), can be assessed via returns on investments using the developed framework.
Strategic Description / RD&T

    
Deployment Plan
As stated in the timeline, deploying the sensing network in the testbed community towards the beginning of the project is a high priority to ensure that the collected data can fully capture correlations to changes in weather, operational, and usage patterns over the course of the project.  Consequently, the deployment will take place during the first quarter from July 2021 through September 2021.  First, the research team will collaborate with our partners at the City of Pittsburgh, Pittsburgh Parks Conservancy, and Metro21 to select the testbed park/community.  Second, the research team will assemble the Urbano nodes in the PI’s lab.  Third, the research team will rigorously test the Urbano nodes in the lab prior to deployment.  Fourth, the team will deploy the Urbano nodes in the testbed community.  In parallel with the assembly process, the team will configure the web application for visualization and descriptive analysis: looking at the data flow between Urbano nodes and the cloud, the Amazon Web Services (AWS) IoT Core will be used to connect Urbano nodes with the cloud, AWS Lambda and AWS S3 will be used to manage computing and storage, respectively, and Plotly will be used to visualize data on a website managed by the research team and accessible to stakeholders.
Expected Outcomes/Impacts
The research team will communicate all intellectual contributions with the scientific community through journal publications, conference presentations/papers, and seminars.  Pending permission from collaborating partners, citizens (in addition to collaborating partners) will be given access to all collected data, which will empower them to engage in how parks are curated and managed.  We expect to attend several stakeholder meetings and listening sessions with the City and Pittsburgh Parks Conservancy.  We will report our findings via presentations to City and Pittsburgh Parks Conservancy over the course of the project.  Proof of concept carried out in a Pittsburgh community will allow the project team to summarize findings in a final report and present recommendations to collaborating partners and stakeholders (i.e., City of Pittsburgh and Pittsburgh Parks Conservancy).  The team will demonstrate to the City how their recommendations, as well as the public’s recommendations (using, for example, the open-source Kurb platform [13] and OpenStreetsPGH [12]), can be assessed via returns on investments using the developed framework (i.e., investments that lower the system-level probability of failure subject to cost constraints).
Expected Outputs

    
TRID


    

Individuals Involved

Email Name Affiliation Role Position
sbenicky@andrew.cmu.edu Benicky, Sheryl Carnegie Mellon University Other Other
kaflanig@cmu.edu Flanigan, Katherine Carnegie Mellon University PI Faculty - Untenured, Tenure Track
lgraff@andrew.cmu.edu Graff, Lindsay Carnegie Mellon University Other Student - PhD
kh3m@andrew.cmu.edu Lightman, Karen Carnegie Mellon University Other Other

Budget

Amount of UTC Funds Awarded
$75654.00
Total Project Budget (from all funding sources)
$151308.00

Documents

Type Name Uploaded
Data Management Plan DataManagementPlan_Mwj3Tao.pdf Dec. 16, 2020, 4:26 p.m.
Presentation Flanigan_ppt.pptx March 19, 2021, 9:26 a.m.
Presentation LinFlaniganBergesPoster.pdf Sept. 30, 2021, 4:41 p.m.
Progress Report 366_Progress_Report_2021-09-30 Sept. 30, 2021, 4:41 p.m.
Progress Report 366_Progress_Report_2022-03-30 March 30, 2022, 4:59 p.m.
Final Report 366_-_Final_Report.pdf Aug. 12, 2022, 12:24 p.m.
Publication Flanigan_paper.pdf Oct. 3, 2022, 12:19 p.m.
Progress Report 366_Progress_Report_2022-09-30 Oct. 3, 2022, 12:19 p.m.

Match Sources

No match sources!

Partners

Name Type
City of Pittsburgh Deployment Partner Deployment Partner
Pittsburgh Parks Conservancy Deployment Partner Deployment Partner
Metro21 Deployment Partner Deployment Partner
University of Pittsburgh Deployment Partner Deployment Partner
University of Michigan Deployment Partner Deployment Partner