Network Analysis

Analyzing Bus Routes to high Schools for Morongo Unified School District

By Matthew Martin

Problem Statement

This project aims to analyze school bus routing to two high schools in the Morongo Unified School District. While one of the smaller than most districts in California (compared to nearby districts like San Bernardino City USD), the area of the district is very large, covering over 1,400 square miles. (Morongo). This makes it challenging for the district to ensure transportation for every student. Using network analysis, I seek to answer the following questions:

What is the optimum allocation for students in the area for each high school?
What are the optimized routes for buses to take to reach each stop?
What areas in the Morongo Unified School District Boundaries are within 30 minutes in walking time to each bus stop?

Data

The Morongo Unified School District is located in San Bernardino County, California. It is part of the 'High Desert" or the western Mojave Desert. It serves the city of Twentynine Palms, the town of Yucca Valley, and the unincorporated areas of Homestead Valley, Joshua Tree, Morongo Valley, and Wonder Valley, as well as the Marine Corps Air Ground Combat Center in Twentynine Palms.

This project will focus on the two comprehensive high schools in the district - Yucca Valley High School and Twentynine Palms High School. The Morongo Unified District Boundary has been clipped from the "California School District Areas 2023-24" dataset from the California Department of Education. The school sites data was used from the "San Bernardino County School Sites" dataset and the streets data from the "SBC Streets" dataset both from San Bernardino County. The bus stops themselves come from the Morongo Unified School District's transportation services website, and geocoded to create a dataset of points.

Methods

This project uses the NAD 1983 State Plane Coordinate System. The State Plane Coordinate system is preferred due to its "high level of accuracy is achieved through the use of relatively small zones" (USGS), and works well for mapping regional areas. Since this project focuses on the network of roads in a small area of California, I chose for minimal distortion and the higher accuracy that come with it.

This project assumes two travel modes: Bus Routing (the time it takes a bus to drive through each road) and Walking (the time it takes a student to walk). In order to find the the driving time for the busses, I estimated the speed limit of a bus for each type of road, divided the length of the road in miles by the speed limit. Then, I multiplied the result by 60 to give a time in minutes for the bus to travel that road. For Walking, I used the average walking speed found from a study (Alves et al.), where an individual under the age of 30 is about 3 mph. I divided the length in miles of each road by 3 and multiplied the result by 60. Cutoffs for each study have been set at different amounts, depending on the question being asked. These values will help calculate the time needed for the busses to stay on schedule and get to each school in time, as well as the time it takes students who do not have other types of transportation to walk to a bus stop. This project also assumes that all roads on land are accounted for in the dataset and that the buses can traverse all types of roads.

What is the optimum allocation for students in the area for each school?

In order to set up the optimum allocation to schools, I first create a Polygon around the roads of Morongo USD Bounds. This will remove all the areas that are not used for housing like federally owned land. Then, I created random points to act as a sample set of students needing to attend one of the two high schools.

Next, I used Network Analyst to create a Location-Allocation Analysis Layer, where I set the travel mode to BusRouting, set Facilities to use the two high schools, set demand points to the random sample of students, and set the problem type to Maximize Attendance.

What are the optimized routes for buses to take to reach each stop?

Next, I again used network analysis with the Optimize Routes for Fleet Management tool. This creates a Vehicle Routing Problem layer. For this layer, I set Orders to be each bus stop, Depots to the two high schools, and routes to be each bus in the fleet organized by route number. Using the StartTime and EndTime in the high school's table and the ServiceTime in the stops layer, Accurate timing can be used to ensure that if leaving the school yards by 5:30 AM, each route will make it to the schools by 7:10 AM (the time both schools start classes).

What areas in the Morongo Unified Boundaries are within 30 Minutes walking time to each bus stop?

Lastly, I used the Service Area tool in network analysis to find what parts of the area are accessible within 30 minutes of walking time to the nearest bus stop. here, I set the travel mode to Walking, using the time it takes for average teens to walk each street. Then I used my BusStops point layer as the facilities, set the direction to "toward facilities" and the service area cutoff to 30 minutes. This will give me a layer showing a polygon around each bus stop that is walkable to within 30 minutes away.

Results

What is the optimum allocation for students in the area for each school?

Here, you can see the optimum allocation of students to help determine which school they should go, based on geographic location and time it takes to drive to each school. This uses a random sampling of 100 data points that represent a prospective student. This shows that if a student lives to the west of the 116°15'W line, then they should to to Yucca Valley high School. If they live to the east of that line, then they should go to Twentynine Palms High School.

What are the optimized routes for buses to take to reach each stop?

This map result shows the optimal routes that the network analyst tool chose for the school busses to pick up students. Routes 1 through 18 represented with circles service Yucca Valley High School on the left, and routes 29 through 39 represented with diamonds service Twentynine Palms High School on the right. It also shows that there is an area in the middle that is not very well serviced. These busses leave the yard at 5:45 AM and return in time for school to start at 7:10 AM. This routing allows busses 14, 15, 17, 18, 38 and 39 to not be used and allows all bus stops to be visited.

What areas in the Morongo Unified Boundaries are within 30 Minutes walking time to each bus stop?

This map shows all the areas (in green) that a student can walk to a bus stop within 30 minutes of walking time. Using an average speed of 3 mph, it shows that some areas in the upper middle part of the boundaries cannot be reached. This means that students would have to walk more than 30 minutes or have a parent drive them to their nearest bus stop. This analysis only assumes that students walk on roads or streets, and do not cut through open land.

Conclusion

The map titled "To Which High School Will Students Go" shows the results of the Location-Allocation tool. To answer the question "What is the optimum allocation for students in the area for each high school?", the optimum allocation for students on the west side of the map will go to Yucca Valley High School, and the students on the east side will go to Twentynine Palms High School. This allocation took into account the drive time it would take to get to each school and the geographic location compared to each school. However, this analysis did not take into account the population maximums for each school, contributing to the limitations on scope for the project. Using the random sampling of points in the city area helped to gain a sense of understanding if looking at random student locations, but did not take into account the population densities for each neighborhood in the area.

The second map shows the optimized routes for buses to take to reach each stop to pick up students and return them to the school sites before the start time of 7:10 AM. This will require the Bus Drivers to leave the schools by 5:45 AM, staying at each stop for five minutes. This routing plan is able to leave 6 busses not used, in order to account for any substitutions, field trips, or sports. Depending on the hours needed for the other schools in the district as well as the amount of drivers available to the district, this could end up saving the district money on wages for the drivers. However, this analysis does not take into account any traffic that may be experienced, especially in the heart of towns and on major roadways. It also does not give any alternated in the case of flooding that the area sometimes experiences. More studies using flood analysis, and creating barriers based on the frequent flood areas could lead to a more robust analysis on the best bus routes despite many different conditions.

Finally, the third map answers the question on what areas in the Morongo Unified School District Boundaries are within 30 minutes in walking time to each bus stop? This is an attempt to understand the accessibility of all students that need to ride the bus to school, and do not have access to a parent driving them. The areas in green are all accessible, but there are still some larger areas that are not. For these areas, students will either need to walk for longer periods, or find another way to reach the bus stop alone. Depending on the time that the busses finally reach each stop, this means that students could need to leave their house by 5:15 Am to get to the bus in time. For some students, this could mean that they may need to wake up before 5:00 AM. This is especially notable since there are many studies that show "that later start times improve teens’ learning and well-being" (APA). If the district does not want to push back their school start times, then understanding the accessibility of the bus routes may help reduce the time needed to get to the school. However, this part of the analysis is limited since it only takes into account that students only walk on designated roadways. There is a lot of open space in and around the city that students can cut through to reach more distance in the 30 minutes. More study on the availability for certain areas could make the conclusions of this study stronger.

References

Alves, F. et al. (2020) ‘Walkability index for Elderly Health: A proposal’, Sustainability, 12(18), p. 7360. doi:10.3390/su12187360.
American Psychological Association. “School Start Times and Sleep.” American Psychological Association (APA), https://www.apa.org/topics/children/school-start-times. Accessed 30 July 2025.
California Department of Education. California School Dashboard: Morongo Unified - 2024 Report. California School Dashboard, https://www.caschooldashboard.org/reports/36677770000000/2024. Accessed 30 July 2025.
California Department of Education. “California School District Areas 2023-24.” ArcGIS Online, https://uagis.maps.arcgis.com/home/item.html?id=cce0dfa63e1f40f89f678cb205fb7168. Accessed 28 July 2025.
Morongo Unified School District. "Transportation." Morongo Unified School District, https://www.morongousd.com/page/transportation. Accessed 28 July 2025.
San Bernardino County. “Centerlines.” San Bernardino County Open Data Hub, https://open.sbcounty.gov/datasets/959001cc5dd0449fbd9b203ad45f5ae5_0/explore. Accessed 28 July 2025.
San Bernardino County. “San Bernardino County School Sites.” ArcGIS Online, [https://uagis.maps.arcgis.com/home/item.html?id=421bf93dadcb421f9a22434a0a4d8500]. Accessed 28 July 2025.
U.S. Geological Survey. “What Is the State Plane Coordinate System? Can GPS Provide Coordinates in These Values?” U.S. Geological Survey, 6 July 2021, https://www.usgs.gov/faqs/what-state-plane-coordinate-system-can-gps-provide-coordinates-these-values. Accessed 30 July 2025

Page updated

Google Sites

Report abuse