Problem Statement

Coffee is one of the most popular beverages in the United States, with roughly two-thirds of adults drinking coffee each day (National Coffee Association, 2025). Dutch Bros is a relatively new coffee chain that has gained significant popularity throughout Arizona and continues to expand into growing communities. Identifying an optimal location for a new Dutch Bros shop requires evaluating multiple spatial factors, including land use, population distribution, transportation access, and competition from existing coffee shops. Suitability analysis can be used to determine the most appropriate location by comparing these criteria through both Boolean and ranked methods. 

Research Question

Where is the most suitable location for a new Dutch Bros coffee shop between Sahuarita and Marana in Pima County, based on commercial land use, adult population, proximity to major roads, and distance from existing coffee shops?

Data

Study Area

The study area is located in Pima County in southern Arizona, specifically within the corridor between Sahuarita and Marana, where a significant portion of recent population growth and residential development has occurred. This area is also experiencing expanding commercial development, including the addition of new Dutch Bros coffee shop locations, with further expansion likely due to increasing demand. The combination of growth, accessibility, and commercial activity makes this area an appropriate focus for evaluating potential locations for a new coffee shop.

Data

Existing Dutch Bros coffee shop addresses were geocoded into latitude and longitude coordinates using the MapQuest converter and were cross-checked to verify accuracy. Zoning, census, and street network data were collected from Pima County government sources, while population areas with missing data were excluded from the study. For consistency, all datasets were clipped to the study area boundary prior to analysis.

Name of Dataset: Census Blocks - 2020 

Publication/Update Date: June 13, 2022 

Geometry Type: Polygon Feature Layer 

Owner: Pima County GIS 

Description: 2020 Census Block boundaries as supplied by the U.S. Census Bureau. Census Blocks are the smallest unit used by the Census Bureau for demographic tabulation. This dataset was used to represent the population of adults (ages ≥18).

Name of Dataset: Street Network

Publication/Update Date: February 24, 2021 

Geometry Type: Line Feature Layer 

Owner: Pima County GIS 

Description: This dataset represents all streets in Pima County using a single-line format. Major roads were identified from this layer.

Name of Dataset: Zoning - City of Tucson 

Publication/Update Date: February 24, 2021 

Geometry Type: Polygon Feature Layer 

Owner: Pima County GIS 

Description: Zoning data for the City of Tucson. This layer was used to identify commercial zones.

Name of Dataset: Zoning - Pima County 

Publication/Update Date: February 24, 2021 

Geometry Type: Polygon Feature Layer 

Owner: Pima County GIS 

Description: Zoning data for Pima County. This layer was used to identify commercial zones. 

Name of Dataset: Zoning - Town of Marana 

Publication/Update Date: February 24, 2021 

Geometry Type: Polygon Feature Layer 

Owner: Pima County GIS 

Description: Zoning data for the Town of Marana. This layer was used to identify commercial zones. 

Name of Dataset: Zoning - Town of Oro Valley 

Publication/Update Date: February 24, 2021 

Geometry Type: Polygon Feature Layer 

Owner: Pima County GIS 

Description: Zoning data for the Town of Oro Valley. This layer was used to identify commercial zones.

Name of Dataset: Zoning - Town of Sahuarita 

Publication/Update Date: February 24, 2021 

Geometry Type: Polygon Feature Layer 

Owner: Pima County GIS 

Description: Zoning data for the Town of Sahuarita. This layer was used to identify commercial zones. 

Methods

The datasets for this project were projected using the NAD 1983 HARN StatePlane Arizona Central FIPS 0202 (Intl Feet) (WKID 2868) coordinate system. This projection was selected because it is commonly used for local datasets in the Tucson region and was the original coordinate system of the Pima County data used in this study. Using a local projected coordinate system minimizes spatial distortion and improves accuracy in distance-based analyses, such as buffering and proximity calculations.

Four criteria were used to determine suitable locations for a new coffee shop. First, new locations must be sufficiently far from existing coffee shops to reduce market competition. Second, census block groups must have an adult population (age ≥18) of at least 300 to target areas with higher potential demand. Third, locations must fall within commercial zoning districts (CB-2 or C-2), which represent general business areas suitable for drive-through coffee shops (Pinal County, Arizona, n.d.). Finally, sites must be located near major roads to ensure accessibility and visibility. These criteria were selected to balance demand, accessibility, zoning requirements, and competition.

Boolean Method

The Boolean suitability analysis identifies areas that meet all criteria simultaneously using a binary approach in which locations are classified as either suitable or not suitable. Each dataset was first processed individually: existing coffee shops were buffered by 1 mile and excluded, census blocks with populations of at least 300 adults were selected, commercial zones (CB-2 and C-2) were isolated, and major roads were buffered by 0.25 miles. The resulting layers were then combined using the Intersect tool to identify only those locations that satisfied all conditions, producing a final map of suitable areas.

Ranked Method

The ranked suitability analysis assigns varying levels of suitability to each criterion rather than using a strict binary approach. Distances from existing coffee shops and major roads were classified using multiple ring buffers (Table A and D, respectively), while adult population values were ranked based on predefined ranges (Table B). Commercial zones were treated as highly suitable, while non-commercial areas were excluded (Table C). Each criterion was assigned a weight to reflect its relative importance: 50% for population, 30% for proximity to major roads, and 20% for distance from existing locations. All layers were then combined using a weighted overlay. This method allows for clearer comparison among locations with different levels of suitability.

Assumptions

Several assumptions were made to guide this analysis. First, a minimum adult population threshold of 300 was selected to represent areas with a sufficient customer base, although this value is somewhat arbitrary and may vary in reality. In addition, a 1-mile buffer from existing coffee shops was used to reduce competition, assuming that closer proximity would negatively affect business performance.

Commercial zoning classifications (CB-2 and C-2) were also assumed to be consistent indicators of general business areas across jurisdictions, despite possible variations in zoning definitions. In the ranked analysis, weights were assigned to emphasize population (50%) over proximity to major roads (30%) and distance from competitors (20%), based on the assumption that customer demand is the most important factor. Although these assumptions simplify real-world conditions, they provide a reasonable framework for identifying suitable locations.

Results

Boolean Suitability Analysis

The Boolean suitability analysis identified 123 suitable locations that met all four criteria. These locations were primarily clustered around the outer edges of existing Dutch Bros buffers, indicating that many optimal sites are located just beyond the 1-mile exclusion zones. Several additional suitable locations appeared in less saturated areas farther from existing shops, suggesting possible opportunities for expansion into underserved regions.

Ranked Suitability Analysis

The ranked suitability analysis produced 286 potential locations after areas with missing population data were removed. Overall, suitability was widely distributed across the study area, with most locations falling within moderate to low suitability categories. Of these, 52 locations were classified as highly suitable (rank 1), with the greatest concentrations occurring along major road corridors and forming linear patterns of high suitability. These highly suitable sites were generally located away from existing Dutch Bros locations, indicating strong potential for expansion into currently underserved areas.

Conclusion

The Boolean and ranked suitability analyses produced largely consistent results in identifying high-suitability locations for a new Dutch Bros coffee shop. This similarity is expected, as both methods were based on the same core criteria, including distance from existing locations, minimum adult population, commercial zoning, and proximity to major roads. In both analyses, the most suitable areas were typically located along major transportation corridors and just outside the 1-mile buffer zones surrounding existing shops. These patterns indicate that accessibility and population density are key drivers in determining optimal locations.

While the Boolean method strictly identified locations that met all criteria simultaneously, the ranked suitability analysis provided a more detailed representation of suitability across the study area. The ranked method revealed a broader distribution of potential sites, including both highly suitable and moderately suitable locations, allowing for greater flexibility in decision-making. Although many high-suitability locations clustered near existing shops, both methods also identified several areas farther from current locations, suggesting opportunities for expansion into underserved regions.

Overall, the two methodologies generally agree on where the most suitable locations are, but the ranked analysis offers additional insight by differentiating between varying levels of suitability. This makes it more useful for prioritizing among multiple candidate sites, while the Boolean method is more effective for identifying strictly acceptable locations.

Limitations

Several limitations may have affected the results of this analysis. First, most of the datasets used were not fully up to date, with zoning and other spatial data dating from approximately 2021 and census data from 2020. Changes in population growth, land use, or new business development since that time may have influenced the accuracy of the results. In addition, the locations of existing Dutch Bros shops were manually collected, which may have introduced minor positional inaccuracies.

Key parameters, such as buffer distances and population thresholds, were also selected based on general assumptions rather than empirical data. For example, the 1-mile exclusion zone and minimum adult population of 300 may not fully reflect real-world market behavior or business strategy. These assumptions simplify a complex site selection process and may have influenced which locations were identified as suitable. Future analyses could be improved by incorporating more recent data and refining the criteria using industry standards or consumer behavior data.

References

Pinal County, Arizona. (n.d.). Development services code and floodplain management: Chapter 2.95 CB-2 general business zone. Municode Library. https://library.municode.com/az/pinal_county/codes/development_services_code_and_floodplain_ management?nodeId=DEVELOPMENT_SERVICES_CODE_TIT2ZO_CH2.95GEBUZO 

Pima County. (n.d.). Street network [Data set]. ArcGIS Hub. Retrieved April 8, 2026, from https://gisopendata.pima.gov/datasets/PimaMaps::street-network/about

Pima County. (n.d.). Census blocks 2020 [Data set]. ArcGIS Hub. Retrieved April 8, 2026, from https://gisopendata.pima.gov/datasets/PimaMaps::census-blocks-2020/about

MapQuest. (n.d.). Latitude/longitude finder [Web tool]. Retrieved April 8, 2026, from https://developer.mapquest.com/documentation/tools/latitude-longitude-finder/

National Coffee Association. (2025, April 15). More Americans drink coffee each day than any other beverage, bottled water back in second place. https://www.ncausa.org/Newsroom/More-Americans-Drink-Coffee-Each-Day-Than-Any-Other-Beverage- Bottled-Water-Back-in-Second-Place 

Zoning datasets

Pima County. (n.d.). Zoning – City of Tucson [Data set]. ArcGIS Hub. Retrieved April 9, 2026, from https://hub.arcgis.com/datasets/PimaMaps::zoning-city-of-tucson/about

Pima County. (n.d.). Zoning – Pima County [Data set]. ArcGIS Hub. Retrieved April 9, 2026, from https://hub.arcgis.com/datasets/PimaMaps::zoning-pima-county/about

Pima County. (n.d.). Zoning – Town of Marana [Data set]. ArcGIS Hub. Retrieved April 9, 2026, from https://gisopendata.pima.gov/datasets/PimaMaps::zoning-town-of-marana/about

Pima County. (n.d.). Zoning – Town of Oro Valley [Data set]. ArcGIS Hub. Retrieved April 9, 2026, from https://gisopendata.pima.gov/datasets/PimaMaps::zoning-town-of-oro-valley/about

Pima County. (n.d.). Zoning – Town of Sahuarita [Data set]. ArcGIS Hub. Retrieved April 9, 2026, from https://gisopendata.pima.gov/datasets/zoning-town-of-sahuarita/about