Skip to content

Which is More Fun: Weekdays or Weekends? A Spatiotemporal Analysis of GPS Data

License

Notifications You must be signed in to change notification settings

ayushdabra/spatiotemporal-gps-data-analysis

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Which is More Fun: Weekdays or Weekends? A Spatiotemporal Analysis of GPS Data

Tech Stack

Problem Statement

Urban spaces are shaped by people's spatiotemporal activities. Analyzing the demands of various social groups and individual differences is necessary in order to develop policies and commercial centers that are based on the needs of the people. Historically mobility patterns and routes were recorded using traditional techniques like travel diaries to obtain spatiotemporal information about individuals. However, these techniques have their own limitations in terms of accuracy and reliability. Modern geo-information technologies, such as GPS-enabled smartphones, can now be used to gather precise and reliable data. The ability to gather a large amount of information while also being precisely localized is a key advantage of such technologies. GPS data can be used to track user mobility and provide valuable insights into the user's preferences and lifestyle habits. These insights can be used to offer location-based services on a large scale, such as assisting cities in deciding where bus routes should be formed to benefit the largest number of people, notifying drivers about slow traffic, assisting businesses in locating high-traffic areas, providing advertisements about businesses along one’s intended route, etc.

The IISER Bhopal campus in Bhopal, India, serves as the region of interest for the study. This study aims to analyze the spatiotemporal mobility patterns on a college campus through smartphone-based GPS monitoring over a week to identify and assess a single student's key daily actions, priorities, and unusual behaviors and compare mobility patterns on weekdays and weekends. The spatial mobility patterns are analyzed using simple mobility indicators such as speed, time spent, and mode of transportation, as well as identifying regions of concentration of specific trajectories within the study area. The study's findings could be valuable for monitoring campus security, finding hot spots on campus, improving campus amenities, and learning more about students' mobility patterns.

Objectives

This study illustrates the idea of GPS-based intelligence by monitoring a single university student's spatiotemporal attributes and behavior over a week using a smartphone-based GPS receiver and displaying the data on a geographic information system (GIS). In particular, this study aims to comprehend the following behavioral characteristics displayed by the student:

  • Analyze the student's daily schedule for a week.
  • Analyze sudden variations in routine and behavior over the weekends.
  • Find the student's preferred route, especially on working weekdays.
  • Identify the student's mode of transportation.

Methodology

In this study, a 21-year-old college student's daily movements were monitored for one week, from Sunday, October 09, 2022, to Saturday, October 15, 2022. The participant resided in Bhopal, India, throughout the course of the study and carried a smartphone with the GPS Logger [1] application installed on it. The application was configured to collect location information each second. The following guidelines were followed for data collection:

  • As soon as the student leaves his residence (i.e., hostel) at the beginning of each day, the GPS Logger application is started. The student turns off the application when he returns to the hostel at the end of the day.
  • While walking and traveling, the student always kept the smartphone-based GPS device in his pocket or bag.
  • When the student spent more time inside buildings than outside, it was most difficult to measure its movement. Due to the "urban canyon effect," [2] signal loss and searching for a signal frequently happened inside buildings. As a result, the automatic search for a signal started again, often generating several incorrect position coordinates.

At the end of each day, the data was imported into the QGIS software and visualized before being used for calculations and data analysis. Additionally, the Time Manager [3] plugin is used to identify the daily commute route and examine the mobility pattern. The kernel density estimation technique is used to identify hotspots in the form of heatmaps by estimating the density based on the number of points in a particular location, with larger numbers of clustered points translating into larger values. In an effort to determine the method of transportation for a specific timeframe, the change in speed over time is also analyzed.

Results

Figure 1 shows the student's commute path on Sunday, October 09, 2022, while Figure 2 shows the student's travel route from Monday, October 10, 2022, to Saturday, October 15, 2022. Figure 1 makes apparent that the student visited Aashima Anupama Mall outside of the IISER Bhopal campus. While Figure 2 shows that the mobility pattern is monotonous during the weekdays. It's interesting to observe that the student's mobility pattern on Saturday, shown in Figure 2(f), is considerably different from previous days because, in contrast to other days, the student never left the hostel area. It is quite apparent that the student mobility patterns are pretty similar during the weekdays and drastically different during the weekend.

Figure 1. Map showing the GPS travel trajectories of a college student on Sunday, October 09, 2022.

Figure 2. Maps showing the GPS travel trajectories of a university student on (a) Monday, October 10, 2022 (b) Tuesday, October 11, 2022, (c) Wednesday, October 12, 2022 (d) Thursday, October 13, 2022, (e) Friday, October 14, 2022 (f) Saturday, October 15, 2022.

Figure 3. GPS track heatmap showing the hotspots on Sunday, October 09, 2022.

Figure 4. GPS track heatmaps showing the hotspots on (a) Monday, October 10, 2022 (b) Tuesday, October 11, 2022, (c) Wednesday, October 12, 2022 (d) Thursday, October 13, 2022, (e) Friday, October 14, 2022 (f) Saturday, October 15, 2022.

Figures 3 and 4 depict the heatmaps generated from the GPS data, which provide a more in-depth look at the student's mobility by visually displaying how much time the student spent in each hotspot. The heatmaps depicted in Figures 3 and 4 were visualized using the Time Manager plugin in QGIS. Figures 3 and 4 illustrate that the student stays in Hostel 7 and eats at Mess 5. On Sunday (see Figure 3), he boarded a bus at the college's main gate and traveled to the Aashima Anupama Mall, where he spent a considerable amount of time, most likely watching a movie. He then encountered some traffic while traveling to Lalghati after waiting for the bus at the bus stop near the mall. He presumably spent some time in a restaurant in Lalghati before moving on to an ice cream parlor and spending some time there as well. After that, he went to the Lalghati bus stop to wait for the bus to make his way back to the college's campus.

Now, if we compare the student's mobility patterns on Sunday and during the weekdays, we will see a clear distinction between the two. The student adheres to a very rigid travel schedule that is mostly focused on getting to and from the hostel, mess, lecture hall complex (LHC), and main building. Over the course of the study, the student's mobility pattern is rather consistent and monotonous from Monday to Friday. It's also interesting to note that the mobility pattern again drastically changed on Saturday when the student spent the entire day in the hostel area and only left the hostel to dine in the mess. Figures 4 (c), (e), and (f) highlight some minor details, showing that the student spent some time at the Ideation Hut and on benches near the cricket field's boundary, perhaps for socializing and interacting with other students. The student's one-way and two-way routes are depicted on the mobility graph for weekdays and weekends in Figure 5.

Figure 5. Graph illustrating the difference between weekdays and weekends in terms of mobility.

The most used path is determined by first combining all the tracks into one shapefile and using the “Line Density” tool from QGIS. Figure 6 depicts the student's preferred routes throughout the course of the study. The student could be seen to routinely take two routes, one from Hostel 7 to Mess 5 and another from Hostel 7 to the lecture hall complex (LHC). Additionally, it can be seen that the student hardly ever uses the road that runs along the hillside and only chooses to use it on weekends when he leaves campus.

Figure 6. (a) Combined mobility tracks of the student within the campus, (b) Line density plot showing most used paths by the student within the campus.

Throughout the study, the student's speed is also analyzed, depicted in Table 1 and Figure 7. Figure 7(a) shows that on Sunday, October 9, 2022, when the student was traveling outside the campus, he attained a maximum speed of 80.64 km/h. It could be assumed that the student had used the bus to get around the city. The maximum speed was 28.47 km/h on Monday, October 10, 2022, which is much faster than the maximum speed on other working weekdays. A closer look at Figure 7(b) reveals that, in contrast to other days, the student turned on his GPS Logger app very late, at around 10:45 AM. This, combined with the fact that the student's speed at this time was at its maximum, raises the possibility that the student was running behind schedule for his 11:00 AM class and may have taken an e-rikshaw to get to the lecture hall complex (LHC). Figure 7 also demonstrates periodic gaps in the student's mobility where his speed is zero, indicating periods when he might be in class, working in the lab, or eating in the mess.

Figure 7. Speed vs. time plots for each day.

Table 1. Minimum, maximum, and average speeds for each day.

Day Minimum
Speed (km/h)
Maximum
Speed (km/h)
Average
Speed (km/h)
Sunday, October 09, 2022 0.00 80.64 7.31
Monday, October 10, 2022 0.00 28.47 2.77
Tuesday, October 11, 2022 0.00 10.90 2.15
Wednesday, October 12, 2022 0.00 8.35 0.82
Thursday, October 13, 2022 0.00 14.94 1.23
Friday, October 14, 2022 0.00 17.92 1.75
Saturday, October 15, 2022 0.00 8.28 0.87

Discussions and Suggestions

This study attempted to map the spatiotemporal mobility patterns of a single college student and analyze his behaviors and priorities over a period of one week. The study provided extremely thorough insights into the student’s daily mobility patterns and how it varies over time, particularly on weekends when he takes a break from the monotonous routine to socialize, enjoy, and explore the city. It also revealed a plethora of information about the student and introduced us to a wide range of open research problems that can be further investigated in future studies. One such problem is the early detection of depression in college students using smartphone-based GPS data. Sohrab Saeb et al. leveraged mobile phone sensor data, including GPS and phone usage, and provided behavioral markers that were strongly related to depression [4]. Along similar lines, patterns in mobility behavior can be used to measure depression passively. The following factors can be weighted together to calculate the degree of depression:

  • Confinement: It can be determined by counting the number of times a particular student leaves the hostel to go to classes or labs, as well as the total length of time that student spends there. Also, examining whether or not a student participates in sports makes it possible to determine the level of physical activity.
  • Loss of appetite: It can be easily measured by analyzing the mobility pattern of the student around the dining halls, canteens, and other eateries within the campus. A relationship between the student and lack of appetite can be established if the student is skipping meals.
  • Lack of interest in studies: The number of times a student attends class each week can be calculated using GPS data. It can be presumed that a student who skips classes is dealing with a problem that is causing them to lose interest in their academics.
  • Disturbed sleep: If a student leaves the hostel late at night or is awake during the night, sleeping disorders like insomnia can be detected from mobility patterns by analyzing and quantifying the frequency of these kinds of incidents.

Data Availability

The datasets generated during and/or analyzed during the current study are available from the author on reasonable request.

References

  1. GPS Logger. BasicAirData. Accessed: Oct. 20, 2022. [Android]. Available: https://play.google.com/store/apps/details?id=eu.basicairdata.graziano.gpslogger&hl=en_IN &gl=US
  2. H. Gong, C. Chen, E. Bialostozky, and C. T. Lawson, “A GPS/GIS method for travel mode detection in New York City,” Comput. Environ. Urban Syst., vol. 36, no. 2, pp. 131–139, Mar. 2012, doi: 10.1016/j.compenvurbsys.2011.05.003.
  3. Anita Graser, Karolina Alexiou, and Seyed Javad Adabikhsoh, TimeManager. 2011. Accessed: Oct. 21, 2022. [Online]. Available: https://plugins.qgis.org/plugins/timemanager/
  4. S. Saeb et al., “Mobile Phone Sensor Correlates of Depressive Symptom Severity in Daily-Life Behavior: An Exploratory Study,” J. Med. Internet Res., vol. 17, no. 7, p. e4273, Jul. 2015, doi: 10.2196/jmir.4273.

About

Which is More Fun: Weekdays or Weekends? A Spatiotemporal Analysis of GPS Data

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published