# Sample: Banking, Mumbai: report

Project: Sample: Banking, Mumbai  
Goal: Describe the project Sample: Banking, Mumbai: run the analyses that tell the most about its data, draw every result, and write a cited report.  
Run: cd8e89bb-fb40-4fa7-a600-6d9a7cc0c9e5 (running, 11 agent run(s))  
Generated: 2026-10-18T06:25:52Z, version 1  
Model cost: USD 0.0000 over 0 model call(s)  

## Data used

| Data | Kind | Rows | Time span | Summary | Source |
| --- | --- | --- | --- | --- | --- |
| Branches | vector, Point | 30 |  | 30 point features; 'type' has 3 values (top: full service 16, digital 8, business 6); 'staff' 6 to 38 (mean 20.37); 'deposits_inr_cr' 171 to 1,143 (mean 517.5); named by 'name' (e.g. Canal Side branch, Station Quarter branch, Market Hill branch); extent 72.8294,19.0285 to 72.9005,19.0885. Point locations of Branches by type. | [90] |
| ATMs | vector, Point | 140 |  | 140 point features; 'location_type' has 4 values (top: station 37, branch 36, mall 34); 'uptime_pct' 88.1 to 99.8 (mean 93.51); extent 72.8284,19.0257 to 72.9010,19.0951. Point locations of ATMs by location_type. | [91] |
| Service areas | vector, Polygon | 24 |  | 24 polygon (area) features; 'population' 16,226 to 160,336 (mean 8.265e+04); 'median_income_inr' 23,680 to 83,761 (mean 5.115e+04); named by 'name' (e.g. Kingsmead, Larchfield, Maple Cross); total area 60.53 km2; extent 72.8270,19.0236 to 72.9032,19.0966. Areas (Service areas) with population to compare and map. | [92] |
| New customer accounts | vector, Point | 1,971 | 2025-04-01 to 2026-09-30 | 1,971 point features; dated by 'opened_on' from 2025-04-01 to 2026-09-30; 'product' has 4 values (top: current 529, loan 521, salary 463); 'channel' has 3 values (top: app 951, branch 818, agent 202); named by 'account_ref' (e.g. AC000001, AC000002, AC000003); extent 72.8271,19.0243 to 72.9024,19.0957. Dated point events by product: where and when New customer accounts happen. | [2] |
| ATM faults | vector, Point | 650 | 2026-04-01 to 2026-09-30 | 650 point features; dated by 'reported_at' from 2026-04-01 to 2026-09-30; 'fault' has 4 values (top: cash out 173, network 172, card reader 160); 'downtime_h' 0.3 to 28.9 (mean 4.26); extent 72.8284,19.0257 to 72.9010,19.0951. Dated point events by fault: where and when ATM faults happen. | [3] |
| Business clients | vector, Point | 259 |  | 259 point features; 'sector' has 5 values (top: retail 62, food 53, logistics 52); 'turnover_inr_cr' 1 to 167.7 (mean 17.91); extent 72.8272,19.0249 to 72.9018,19.0952. Point locations of Business clients by sector. | [93] |
| Competitor branches | vector, Point | 80 |  | 80 point features; 'bank' has 4 values (top: Bank P 26, Bank Q 25, Bank R 15); extent 72.8187,19.0271 to 72.9006,19.0949. Point locations of Competitor branches by bank. | [94] |
| Transactions by area and month | table | 432 | 2025-04-01 to 2026-09-01 | Table of 432 rows and 6 columns without geometry; dated by 'month' from 2025-04-01 to 2026-09-01; 'card_txn' 6,490 to 124,927 (mean 3.929e+04); 'upi_txn' 29,206 to 753,312 (mean 2.088e+05); 'cash_withdrawals_lakh' 10.9 to 128.3 (mean 60.79). A table of Transactions by area and month without locations. | [4] |
| Branch targets | table | 90 |  | Table of 90 rows and 5 columns without geometry; 'quarter' has 3 values (top: 2026-Q1 30, 2026-Q2 30, 2026-Q3 30); 'deposits_target_cr' 31 to 109 (mean 60.63); 'loans_target_cr' 17 to 78 (mean 37.02); 'achieved_pct' 75.2 to 124.8 (mean 102.8). A table of Branch targets without locations. | [95] |
| Suburban rail lines | vector, LineString | 3 |  | 3 line features; 'daily_riders' 759,730 to 1,011,722 (mean 8.475e+05); named by 'name' (e.g. Western line, Central line, Harbor link); total length 22.43 km; extent 72.8434,19.0237 to 72.9029,19.0963. Lines (Suburban rail lines) such as routes or networks. | [96] |
| Branch network strategy memo | document |  |  | 1-page md document 'Branch network strategy memo'; language en; mentions 3 place(s): 19.08544, 72.87250 (1), Canal Side branch (1), Station Quarter branch (1). A document about 19.08544, 72.87250, Canal Side branch, Station Quarter branch. | [97] |
| Sample: Banking, Mumbai | record |  |  |  | [98] |
| Emerging hot spots: ATM faults | layer, Polygon | 84 |  |  | [99] |
| Emerging hot spots: New customer accounts | layer, Polygon | 106 |  |  | [100] |
| Summarize within: Service areas | layer, Polygon | 24 |  |  | [101] |
| Summarize within: Service areas | layer, Polygon | 24 |  |  | [102] |
| Summarize within: Service areas | layer, Polygon | 24 |  |  | [103] |
| Hot spot analysis: ATM faults | layer, Polygon | 240 |  |  | [104] |
| Hot spot analysis: ATMs | layer, Polygon | 68 |  |  | [105] |
| Hot spot analysis: Branches | layer, Polygon | 16 |  |  | [106] |
| Calculate density: ATM faults | layer, Polygon | 1,893 |  |  | [107] |
| Calculate density: ATMs | layer, Polygon | 459 |  |  | [108] |
| Calculate density: Branches | layer, Polygon | 88 |  |  | [109] |
| Find point clusters: ATM faults | layer, Point | 650 |  |  | [110] |
| Find point clusters: cluster hulls | layer, Polygon | 14 |  |  | [111] |
| Find point clusters: ATMs | layer, Point | 140 |  |  | [112] |
| Find point clusters: cluster hulls | layer, Polygon | 1 |  |  | [113] |
| Find point clusters: Branches | layer, Point | 30 |  |  | [114] |
| Find point clusters: cluster hulls | layer, Polygon | 1 |  |  | [115] |
| Find outliers: Service areas | layer, Polygon | 24 |  |  | [116] |
| Repeat and near repeat classification: ATM faults | layer, Point | 650 |  |  | [117] |

## Timeline

![Timeline: events per month by dataset](charts/24-timeline-events-per-month-by-dataset.png)

*Figure 1. Timeline: events per month by dataset [1]. Per month, 2025-04-01 to 2026-09-01. Source: New customer accounts, ATM faults, Transactions by area and month (3,053 features). The data is in the appendix.*

The timeline counts 3,053 dated events in New customer accounts, ATM faults, Transactions by area and month per month, from April 2025 to September 2026 [1][2][3][4]. By dataset: New customer accounts 1,971; ATM faults 650; Transactions by area and month 432 [1]. The trend is +10.8 events per month; the second half of the months has 1,951 against 1,102 in the first (+77.0%) [1].

**Busiest months**

- August 2026: 293 events [1]
- July 2026: 270 events [1]
- April 2026: 258 events [1]

**Quietest months**

- August 2025: 102 events [1]
- February 2026: 112 events [1]
- May 2025: 113 events [1]

## Space time hot spots

**What this analysis is.** Emerging hot spots: Where hot spots are new, consecutive, intensifying, persistent, diminishing, sporadic, oscillating or historical over time [5]. In ArcGIS it is Emerging Hot Spot Analysis or Create Space Time Cube [5].

**Why it was run on this data.** ATM faults carry dates in 'reported_at': emerging hot spot analysis finds new, intensifying and fading hot spots per week [6].

**What it found.** The process saved a layer named Emerging hot spots: ATM faults, which contained 84 features [7]. Using ATM faults as input [8], the operation completed with an elapsed time of 24 milliseconds [7]. The analysis utilized 650 features and generated a count of 84 results [7].

| Result | Value |
| --- | --- |
| result, count | 84 |
| result, elapsed ms | 24 |
| result, resolution | 8 |
| result, features used | 650 |
| result, time neighbors | 0 |
| billing, cost | 0.1325 |
| billing, list price | 0.1325 |
| billing, upstream cost | 0 |

**Findings**

- **Evidenced:** Space time hot spots on ATM faults: Emerging hot spots: 1 layer saved: Emerging hot spots: ATM faults (84 features) [7][8].
- **Evidenced:** It reported result, count 84; result, elapsed ms 24; result, resolution 8; result, features used 650; result, time neighbors 0; billing, cost 0.1325 [7].

*Tool gp_space_time_hot_spots; parameters {"project": "36cc8b7b-aabc-4b71-981b-a2c53e8f31ad", "features": {"layer": "ATM faults (2bdf3635)"}, "interval": "week", "bin_shape": "h3", "resolution": 8, "time_field": "reported_at"}*

## Space time hot spots

**What this analysis is.** Emerging hot spots: Where hot spots are new, consecutive, intensifying, persistent, diminishing, sporadic, oscillating or historical over time [5]. In ArcGIS it is Emerging Hot Spot Analysis or Create Space Time Cube [5].

**Why it was run on this data.** New customer accounts carry dates in 'opened_on': emerging hot spot analysis finds new, intensifying and fading hot spots per month [9].

**What it found.** Space time hot spots on New customer accounts: Emerging hot spots: 1 layer saved: Emerging hot spots: New customer accounts (106 features) [10][11]. It reported result, count 106; result, elapsed ms 21; result, resolution 8; result, features used 1,971; result, time neighbors 0; billing, cost 0.1986 [10].

| Result | Value |
| --- | --- |
| result, count | 106 |
| result, elapsed ms | 21 |
| result, resolution | 8 |
| result, features used | 1,971 |
| result, time neighbors | 0 |
| billing, cost | 0.1986 |
| billing, list price | 0.1986 |
| billing, upstream cost | 0 |

**Findings**

- **Evidenced:** Space time hot spots on New customer accounts: Emerging hot spots: 1 layer saved: Emerging hot spots: New customer accounts (106 features) [10][11].
- **Evidenced:** It reported result, count 106; result, elapsed ms 21; result, resolution 8; result, features used 1,971; result, time neighbors 0; billing, cost 0.1986 [10].

*Tool gp_space_time_hot_spots; parameters {"project": "36cc8b7b-aabc-4b71-981b-a2c53e8f31ad", "features": {"layer": "New customer accounts (e03b0fae)"}, "interval": "month", "bin_shape": "h3", "resolution": 8, "time_field": "opened_on"}*

## Summarize within

**What this analysis is.** Summarize within: Count and aggregate the features inside each polygon (or bin), proportionally for lines and polygons [12]. In ArcGIS it is Summarize Within or Aggregate Points; in QGIS, native:countpointsinpolygon or native:sumlinelengths [12].

**Why it was run on this data.** Count ATM faults inside each feature of Service areas by fault; the counts map as a choropleth (natural breaks, sequential colors) [13].

**What it found.** The process of summarizing ATM faults using the Service areas feature set retrieved 24 features [14]. This execution involved analyzing the Service areas [15] and clocked an elapsed time of 4 ms [14]. The analysis reported a billing cost and list price of 0.0235, noting an upstream cost of 0 [14].

| Result | Value |
| --- | --- |
| result, count | 24 |
| result, elapsed ms | 4 |
| billing, cost | 0.0235 |
| billing, list price | 0.0235 |
| billing, upstream cost | 0 |

**Findings**

- **Evidenced:** Summarize within on ATM faults, Service areas: Summarize within: 1 layer saved: Summarize within: Service areas (24 features) [8][14][15].
- **Evidenced:** It reported result, count 24; result, elapsed ms 4; billing, cost 0.0235; billing, list price 0.0235; billing, upstream cost 0 [14].

*Tool gp_summarize_within; parameters {"stats": [{"stat": "sum", "field": "downtime_h"}], "project": "36cc8b7b-aabc-4b71-981b-a2c53e8f31ad", "features": {"layer": "ATM faults (2bdf3635)"}, "group_by": "fault", "polygons": {"layer": "Service areas (6344158f)"}}*

## Summarize within

**What this analysis is.** Summarize within: Count and aggregate the features inside each polygon (or bin), proportionally for lines and polygons [12]. In ArcGIS it is Summarize Within or Aggregate Points; in QGIS, native:countpointsinpolygon or native:sumlinelengths [12].

**Why it was run on this data.** Count ATM faults inside each feature of Service areas by fault; the counts map as a choropleth (natural breaks, sequential colors) [13].

**What it found.** The `gp_summarize_within` tool was executed to process Service areas [15] using ATMs [17] as input [16]. This process resulted in a count of 24 for the service areas [16]. Costs associated with this operation included a billing cost and list price of 0.0133 [16]. The upstream cost for the billing was zero [16].

| Result | Value |
| --- | --- |
| result, count | 24 |
| result, elapsed ms | 1 |
| billing, cost | 0.0133 |
| billing, list price | 0.0133 |
| billing, upstream cost | 0 |

**Findings**

- **Evidenced:** Summarize within on ATMs, Service areas: Summarize within: 1 layer saved: Summarize within: Service areas (24 features) [15][16][17].
- **Evidenced:** It reported result, count 24; result, elapsed ms 1; billing, cost 0.0133; billing, list price 0.0133; billing, upstream cost 0 [16].

*Tool gp_summarize_within; parameters {"stats": [{"stat": "sum", "field": "uptime_pct"}], "project": "36cc8b7b-aabc-4b71-981b-a2c53e8f31ad", "features": {"layer": "ATMs (f9f05311)"}, "group_by": "location_type", "polygons": {"layer": "Service areas (6344158f)"}}*

## Summarize within

**What this analysis is.** Summarize within: Count and aggregate the features inside each polygon (or bin), proportionally for lines and polygons [12]. In ArcGIS it is Summarize Within or Aggregate Points; in QGIS, native:countpointsinpolygon or native:sumlinelengths [12].

**Why it was run on this data.** Count ATM faults inside each feature of Service areas by fault; the counts map as a choropleth (natural breaks, sequential colors) [13].

**What it found.** Summarize within on Branches, Service areas: Summarize within: 1 layer saved: Summarize within: Service areas (24 features) [15][18][19]. It reported result, count 24; result, elapsed ms 1; billing, cost 0.0111; billing, list price 0.0111; billing, upstream cost 0 [18].

| Result | Value |
| --- | --- |
| result, count | 24 |
| result, elapsed ms | 1 |
| billing, cost | 0.0111 |
| billing, list price | 0.0111 |
| billing, upstream cost | 0 |

**Findings**

- **Evidenced:** Summarize within on Branches, Service areas: Summarize within: 1 layer saved: Summarize within: Service areas (24 features) [15][18][19].
- **Evidenced:** It reported result, count 24; result, elapsed ms 1; billing, cost 0.0111; billing, list price 0.0111; billing, upstream cost 0 [18].

*Tool gp_summarize_within; parameters {"stats": [{"stat": "sum", "field": "staff"}], "project": "36cc8b7b-aabc-4b71-981b-a2c53e8f31ad", "features": {"layer": "Branches (8f400be9)"}, "group_by": "type", "polygons": {"layer": "Service areas (6344158f)"}}*

## Hot spots

**What this analysis is.** Hot spot analysis: Getis-Ord Gi* hot and cold spots with z-scores, p-values and confidence bins [20]. In ArcGIS it is Find Hot Spots or Hot Spot Analysis (Getis-Ord Gi*); in QGIS, qgis:hotspot (plugin) [20].

**Why it was run on this data.** 650 points: Getis-Ord Gi* tests whether the concentrations in ATM faults are statistically significant [21].

**What it found.** Hot spots on ATM faults: Hot spot analysis: 1 layer saved: Hot spot analysis: ATM faults (240 features) [8][22]. It reported result, bins 240; result, count 240; result, elapsed ms 19; result, cell size m 500; result, incidents used 650; result, distance band m 1,000 [22].

| Result | Value |
| --- | --- |
| result, bins | 240 |
| result, count | 240 |
| result, elapsed ms | 19 |
| result, cell size m | 500 |
| result, incidents used | 650 |
| result, distance band m | 1,000 |
| billing, cost | 0.063 |
| billing, list price | 0.063 |
| billing, upstream cost | 0 |

**Findings**

- **Evidenced:** Hot spots on ATM faults: Hot spot analysis: 1 layer saved: Hot spot analysis: ATM faults (240 features) [8][22].
- **Evidenced:** It reported result, bins 240; result, count 240; result, elapsed ms 19; result, cell size m 500; result, incidents used 650; result, distance band m 1,000 [22].

*Tool gp_hot_spots; parameters {"project": "36cc8b7b-aabc-4b71-981b-a2c53e8f31ad", "features": {"layer": "ATM faults (2bdf3635)"}, "aggregation": "hexagon", "cell_size_m": 500.0, "distance_band_m": 1000.0}*

## Hot spots

**What this analysis is.** Hot spot analysis: Getis-Ord Gi* hot and cold spots with z-scores, p-values and confidence bins [20]. In ArcGIS it is Find Hot Spots or Hot Spot Analysis (Getis-Ord Gi*); in QGIS, qgis:hotspot (plugin) [20].

**Why it was run on this data.** 140 points: Getis-Ord Gi* tests whether the concentrations in ATMs are statistically significant [23].

**What it found.** The hot spot analysis tool processed the ATMs layer to generate one layer containing 68 features [24].

| Result | Value |
| --- | --- |
| result, bins | 68 |
| result, count | 68 |
| result, elapsed ms | 5 |
| result, cell size m | 1,000 |
| result, incidents used | 140 |
| result, distance band m | 2,000 |
| billing, cost | 0.0528 |
| billing, list price | 0.0528 |
| billing, upstream cost | 0 |

**Findings**

- **Evidenced:** Hot spots on ATMs: Hot spot analysis: 1 layer saved: Hot spot analysis: ATMs (68 features) [17][24].
- **Evidenced:** It reported result, bins 68; result, count 68; result, elapsed ms 5; result, cell size m 1,000; result, incidents used 140; result, distance band m 2,000 [24].

*Tool gp_hot_spots; parameters {"project": "36cc8b7b-aabc-4b71-981b-a2c53e8f31ad", "features": {"layer": "ATMs (f9f05311)"}, "aggregation": "hexagon", "cell_size_m": 1000.0, "distance_band_m": 2000.0}*

## Hot spots

**What this analysis is.** Hot spot analysis: Getis-Ord Gi* hot and cold spots with z-scores, p-values and confidence bins [20]. In ArcGIS it is Find Hot Spots or Hot Spot Analysis (Getis-Ord Gi*); in QGIS, qgis:hotspot (plugin) [20].

**Why it was run on this data.** 30 points: Getis-Ord Gi* tests whether the concentrations in Branches are statistically significant [25].

**What it found.** The hot spot analysis processed the Branches layer [19], saving one layer containing 16 features [26]. The analysis utilized 30 incidents and generated 16 bins [26].

| Result | Value |
| --- | --- |
| result, bins | 16 |
| result, count | 16 |
| result, elapsed ms | 2 |
| result, cell size m | 2,000 |
| result, incidents used | 30 |
| result, distance band m | 2,000 |
| billing, cost | 0.0506 |
| billing, list price | 0.0506 |
| billing, upstream cost | 0 |

**Findings**

- **Evidenced:** Hot spots on Branches: Hot spot analysis: 1 layer saved: Hot spot analysis: Branches (16 features) [19][26].
- **Evidenced:** It reported result, bins 16; result, count 16; result, elapsed ms 2; result, cell size m 2,000; result, incidents used 30; result, distance band m 2,000 [26].

*Tool gp_hot_spots; parameters {"project": "36cc8b7b-aabc-4b71-981b-a2c53e8f31ad", "features": {"layer": "Branches (8f400be9)"}, "aggregation": "hexagon", "cell_size_m": 2000.0, "distance_band_m": 2000.0}*

## Calculate density

**What this analysis is.** Calculate density: Kernel density per square kilometer on a grid [27]. In ArcGIS it is Calculate Density or Kernel Density; in QGIS, native:heatmapkerneldensityestimation [27].

**Why it was run on this data.** 650 points in ATM faults: a density surface shows where they concentrate [28].

**What it found.** Calculate density on ATM faults: Calculate density: 1 layer saved: Calculate density: ATM faults (1893 features) [8][29]. It reported result, count 1,893; result, elapsed ms 61; result, bandwidth m 1,000; result, cell size m 200; result, excluded null 0; result, max density per km2 45.95 [29].

| Result | Value |
| --- | --- |
| result, count | 1,893 |
| result, elapsed ms | 61 |
| result, bandwidth m | 1,000 |
| result, cell size m | 200 |
| result, excluded null | 0 |
| result, max density per km2 | 45.95 |
| billing, cost | 0.063 |
| billing, list price | 0.063 |
| billing, upstream cost | 0 |

**Findings**

- **Evidenced:** Calculate density on ATM faults: Calculate density: 1 layer saved: Calculate density: ATM faults (1893 features) [8][29].
- **Evidenced:** It reported result, count 1,893; result, elapsed ms 61; result, bandwidth m 1,000; result, cell size m 200; result, excluded null 0; result, max density per km2 45.95 [29].

*Tool gp_calculate_density; parameters {"kernel": "quartic", "project": "36cc8b7b-aabc-4b71-981b-a2c53e8f31ad", "features": {"layer": "ATM faults (2bdf3635)"}, "bandwidth_m": 1000.0, "cell_size_m": 200.0}*

## Calculate density

**What this analysis is.** Calculate density: Kernel density per square kilometer on a grid [27]. In ArcGIS it is Calculate Density or Kernel Density; in QGIS, native:heatmapkerneldensityestimation [27].

**Why it was run on this data.** 140 points in ATMs: a density surface shows where they concentrate [30].

**What it found.** Calculate density on ATMs: Calculate density: 1 layer saved: Calculate density: ATMs (459 features) [17][31]. It reported result, count 459; result, elapsed ms 15; result, bandwidth m 2,000; result, cell size m 500; result, excluded null 0; result, max density per km2 3.678 [31].

| Result | Value |
| --- | --- |
| result, count | 459 |
| result, elapsed ms | 15 |
| result, bandwidth m | 2,000 |
| result, cell size m | 500 |
| result, excluded null | 0 |
| result, max density per km2 | 3.678 |
| billing, cost | 0.0528 |
| billing, list price | 0.0528 |
| billing, upstream cost | 0 |

**Findings**

- **Evidenced:** Calculate density on ATMs: Calculate density: 1 layer saved: Calculate density: ATMs (459 features) [17][31].
- **Evidenced:** It reported result, count 459; result, elapsed ms 15; result, bandwidth m 2,000; result, cell size m 500; result, excluded null 0; result, max density per km2 3.678 [31].

*Tool gp_calculate_density; parameters {"kernel": "quartic", "project": "36cc8b7b-aabc-4b71-981b-a2c53e8f31ad", "features": {"layer": "ATMs (f9f05311)"}, "bandwidth_m": 2000.0, "cell_size_m": 500.0}*

## Calculate density

**What this analysis is.** Calculate density: Kernel density per square kilometer on a grid [27]. In ArcGIS it is Calculate Density or Kernel Density; in QGIS, native:heatmapkerneldensityestimation [27].

**Why it was run on this data.** 30 points in Branches: a density surface shows where they concentrate [32].

**What it found.** The density calculation successfully processed 88 branches [33]. Parameters used in the calculation included a bandwidth of 2,000 m and a cell size of 1,000 m [33]. The analysis generated a maximum density of 0.9778 per km² [33]. The total billing cost associated with this calculation was 0.0506 [33].

| Result | Value |
| --- | --- |
| result, count | 88 |
| result, elapsed ms | 3 |
| result, bandwidth m | 2,000 |
| result, cell size m | 1,000 |
| result, excluded null | 0 |
| result, max density per km2 | 0.9778 |
| billing, cost | 0.0506 |
| billing, list price | 0.0506 |
| billing, upstream cost | 0 |

**Findings**

- **Evidenced:** Calculate density on Branches: Calculate density: 1 layer saved: Calculate density: Branches (88 features) [19][33].
- **Evidenced:** It reported result, count 88; result, elapsed ms 3; result, bandwidth m 2,000; result, cell size m 1,000; result, excluded null 0; result, max density per km2 0.9778 [33].

*Tool gp_calculate_density; parameters {"kernel": "quartic", "project": "36cc8b7b-aabc-4b71-981b-a2c53e8f31ad", "features": {"layer": "Branches (8f400be9)"}, "bandwidth_m": 2000.0, "cell_size_m": 1000.0}*

## Find point clusters

**What this analysis is.** Find point clusters: Density-based clusters (DBSCAN, HDBSCAN or OPTICS, optionally space-time) with hulls and noise [34]. In ArcGIS it is Find Point Clusters or Density-based Clustering; in QGIS, native:dbscanclustering or native:stdbscanclustering [34].

**Why it was run on this data.** DBSCAN finds dense groups of ATM faults and separates noise [35].

**What it found.** Find point clusters on ATM faults: Find point clusters: 2 layers saved: Find point clusters: ATM faults (650 features); Find point clusters: cluster hulls (14 features) [8][36]. It reported result, count 650; result, noise 232; result, distance m 500; result, elapsed ms 14; result, min points 16; billing, cost 0.063 [36].

| Result | Value |
| --- | --- |
| result, count | 650 |
| result, noise | 232 |
| result, distance m | 500 |
| result, elapsed ms | 14 |
| result, min points | 16 |
| billing, cost | 0.063 |
| billing, list price | 0.063 |
| billing, upstream cost | 0 |

**Findings**

- **Evidenced:** Find point clusters on ATM faults: Find point clusters: 2 layers saved: Find point clusters: ATM faults (650 features); Find point clusters: cluster hulls (14 features) [8][36].
- **Evidenced:** It reported result, count 650; result, noise 232; result, distance m 500; result, elapsed ms 14; result, min points 16; billing, cost 0.063 [36].

*Tool gp_find_point_clusters; parameters {"method": "dbscan", "project": "36cc8b7b-aabc-4b71-981b-a2c53e8f31ad", "features": {"layer": "ATM faults (2bdf3635)"}, "distance_m": 500.0, "min_points": 16}*

## Find point clusters

**What this analysis is.** Find point clusters: Density-based clusters (DBSCAN, HDBSCAN or OPTICS, optionally space-time) with hulls and noise [34]. In ArcGIS it is Find Point Clusters or Density-based Clustering; in QGIS, native:dbscanclustering or native:stdbscanclustering [34].

**Why it was run on this data.** DBSCAN finds dense groups of ATMs and separates noise [37].

**What it found.** Find point clusters on ATMs: Find point clusters: 2 layers saved: Find point clusters: ATMs (140 features); Find point clusters: cluster hulls (1 feature) [17][38]. It reported result, count 140; result, noise 0; result, distance m 1,000; result, elapsed ms 3; result, min points 4; billing, cost 0.0528 [38].

| Result | Value |
| --- | --- |
| result, count | 140 |
| result, noise | 0 |
| result, distance m | 1,000 |
| result, elapsed ms | 3 |
| result, min points | 4 |
| billing, cost | 0.0528 |
| billing, list price | 0.0528 |
| billing, upstream cost | 0 |

**Findings**

- **Evidenced:** Find point clusters on ATMs: Find point clusters: 2 layers saved: Find point clusters: ATMs (140 features); Find point clusters: cluster hulls (1 feature) [17][38].
- **Evidenced:** It reported result, count 140; result, noise 0; result, distance m 1,000; result, elapsed ms 3; result, min points 4; billing, cost 0.0528 [38].

*Tool gp_find_point_clusters; parameters {"method": "dbscan", "project": "36cc8b7b-aabc-4b71-981b-a2c53e8f31ad", "features": {"layer": "ATMs (f9f05311)"}, "distance_m": 1000.0, "min_points": 4}*

## Find point clusters

**What this analysis is.** Find point clusters: Density-based clusters (DBSCAN, HDBSCAN or OPTICS, optionally space-time) with hulls and noise [34]. In ArcGIS it is Find Point Clusters or Density-based Clustering; in QGIS, native:dbscanclustering or native:stdbscanclustering [34].

**Why it was run on this data.** DBSCAN finds dense groups of Branches and separates noise [39].

**What it found.** Find point clusters on Branches: Find point clusters: 2 layers saved: Find point clusters: Branches (30 features); Find point clusters: cluster hulls (1 feature) [19][40]. It reported result, count 30; result, noise 0; result, distance m 2,000; result, elapsed ms 2; result, min points 3; billing, cost 0.0506 [40].

| Result | Value |
| --- | --- |
| result, count | 30 |
| result, noise | 0 |
| result, distance m | 2,000 |
| result, elapsed ms | 2 |
| result, min points | 3 |
| billing, cost | 0.0506 |
| billing, list price | 0.0506 |
| billing, upstream cost | 0 |

**Findings**

- **Evidenced:** Find point clusters on Branches: Find point clusters: 2 layers saved: Find point clusters: Branches (30 features); Find point clusters: cluster hulls (1 feature) [19][40].
- **Evidenced:** It reported result, count 30; result, noise 0; result, distance m 2,000; result, elapsed ms 2; result, min points 3; billing, cost 0.0506 [40].

*Tool gp_find_point_clusters; parameters {"method": "dbscan", "project": "36cc8b7b-aabc-4b71-981b-a2c53e8f31ad", "features": {"layer": "Branches (8f400be9)"}, "distance_m": 2000.0, "min_points": 3}*

## Outliers

**What this analysis is.** Find outliers: Anselin local Moran’s I: HH, LL clusters and HL, LH outliers [41]. In ArcGIS it is Find Outliers or Cluster and Outlier Analysis (Anselin Local Moran’s I) [41].

**Why it was run on this data.** Local Moran's I maps clusters and outliers of 'population' [42].

**What it found.** Outliers on Service areas: Find outliers: 1 layer saved: Find outliers: Service areas (24 features) [15][43]. It reported result, count 24; result, elapsed ms 6; result, permutations 199; result, excluded null 0; result, distance band m 1,844; billing, cost 0.05072 [43].

| Result | Value |
| --- | --- |
| result, count | 24 |
| result, elapsed ms | 6 |
| result, permutations | 199 |
| result, excluded null | 0 |
| result, distance band m | 1,844 |
| billing, cost | 0.05072 |
| billing, list price | 0.05072 |
| billing, upstream cost | 0 |

**Findings**

- **Evidenced:** Outliers on Service areas: Find outliers: 1 layer saved: Find outliers: Service areas (24 features) [15][43].
- **Evidenced:** It reported result, count 24; result, elapsed ms 6; result, permutations 199; result, excluded null 0; result, distance band m 1,844; billing, cost 0.05072 [43].

*Tool gp_outliers; parameters {"field": "population", "project": "36cc8b7b-aabc-4b71-981b-a2c53e8f31ad", "features": {"layer": "Service areas (6344158f)"}, "aggregation": "hexagon"}*

## Near repeat

**What this analysis is.** Repeat and near repeat classification: Classify incidents as originators, repeats (same place) or near repeats within a distance and time window, with a Knox test table [44]. In ArcGIS it is Repeat and Near Repeat Classification or Near Repeat Calculator [44].

**Why it was run on this data.** Near repeat analysis tests whether ATM faults close in space also follow each other within 18 days [45].

**What it found.** Near repeat on ATM faults: Repeat and near repeat classification: 1 layer saved: Repeat and near repeat classification: ATM faults (650 features) [8][46]. It reported result, days 18; result, count 650; result, chains 119; result, distance m 200; result, elapsed ms 23; result, pairs linked 690 [46].

| Result | Value |
| --- | --- |
| result, days | 18 |
| result, count | 650 |
| result, chains | 119 |
| result, distance m | 200 |
| result, elapsed ms | 23 |
| result, pairs linked | 690 |
| billing, cost | 0.0695 |
| billing, list price | 0.0695 |
| billing, upstream cost | 0 |

**Findings**

- **Evidenced:** Near repeat on ATM faults: Repeat and near repeat classification: 1 layer saved: Repeat and near repeat classification: ATM faults (650 features) [8][46].
- **Evidenced:** It reported result, days 18; result, count 650; result, chains 119; result, distance m 200; result, elapsed ms 23; result, pairs linked 690 [46].

*Tool gp_near_repeat; parameters {"days": 18, "project": "36cc8b7b-aabc-4b71-981b-a2c53e8f31ad", "features": {"layer": "ATM faults (2bdf3635)"}, "distance_m": 200.0, "time_field": "reported_at"}*

## Maps and charts

The run also made 20 map image(s) [47][48][49][50][51][52][53][54][55][56][57][58][59][60][61][62][63][64][65][66] and 23 chart(s) [67][68][69][70][71][72][73][74][75][76][77][78][79][80][81][82][83][84][85][86][87][88][89].

![Emerging hot spots: ATM faults](maps/01-emerging-hot-spots-atm-faults.png)

*Figure 2. Emerging hot spots: ATM faults [47]*

![Emerging hot spots: New customer accounts](maps/02-emerging-hot-spots-new-customer-accounts.png)

*Figure 3. Emerging hot spots: New customer accounts [48]*

![Summarize within: Service areas](maps/03-summarize-within-service-areas.png)

*Figure 4. Summarize within: Service areas [49]*

![Summarize within: Service areas](maps/04-summarize-within-service-areas.png)

*Figure 5. Summarize within: Service areas [50]*

![Summarize within: Service areas](maps/05-summarize-within-service-areas.png)

*Figure 6. Summarize within: Service areas [51]*

![Service areas by population](maps/06-service-areas-by-population.png)

*Figure 7. Service areas by population [52]*

![Hot spot analysis: ATM faults](maps/07-hot-spot-analysis-atm-faults.png)

*Figure 8. Hot spot analysis: ATM faults [53]*

![Hot spot analysis: ATMs](maps/08-hot-spot-analysis-atms.png)

*Figure 9. Hot spot analysis: ATMs [54]*

![Hot spot analysis: Branches](maps/09-hot-spot-analysis-branches.png)

*Figure 10. Hot spot analysis: Branches [55]*

![Calculate density: ATM faults](maps/10-calculate-density-atm-faults.png)

*Figure 11. Calculate density: ATM faults [56]*

![Calculate density: ATMs](maps/11-calculate-density-atms.png)

*Figure 12. Calculate density: ATMs [57]*

![Calculate density: Branches](maps/12-calculate-density-branches.png)

*Figure 13. Calculate density: Branches [58]*

![Find point clusters: ATM faults](maps/13-find-point-clusters-atm-faults.png)

*Figure 14. Find point clusters: ATM faults [59]*

![Find point clusters: cluster hulls](maps/14-find-point-clusters-cluster-hulls.png)

*Figure 15. Find point clusters: cluster hulls [60]*

![Find point clusters: ATMs](maps/15-find-point-clusters-atms.png)

*Figure 16. Find point clusters: ATMs [61]*

![Find point clusters: cluster hulls](maps/16-find-point-clusters-cluster-hulls.png)

*Figure 17. Find point clusters: cluster hulls [62]*

![Find point clusters: Branches](maps/17-find-point-clusters-branches.png)

*Figure 18. Find point clusters: Branches [63]*

![Find point clusters: cluster hulls](maps/18-find-point-clusters-cluster-hulls.png)

*Figure 19. Find point clusters: cluster hulls [64]*

![Find outliers: Service areas](maps/19-find-outliers-service-areas.png)

*Figure 20. Find outliers: Service areas [65]*

![Repeat and near repeat classification: ATM faults](maps/20-repeat-and-near-repeat-classification-atm-faults.png)

*Figure 21. Repeat and near repeat classification: ATM faults [66]*

**What the map shows.** A highly defined cluster of emerging hot spots is visible in the upper-center portion of the map, indicating concentrated areas for new customer accounts. Within this core area, the color coding reveals a [48]

*Inferred: read from the image by gemma4:e4b, not evidence.*

![ATM faults over time (reported_at)](charts/01-atm-faults-over-time-reported-at.png)

*Figure 22. ATM faults over time (reported_at) [67]. Per week, 2026-03-30 to 2026-09-28. Source: ATM faults (650 features). The data is in the appendix.*

![Transactions by area and month over time (month)](charts/02-transactions-by-area-and-month-over-time-month.png)

*Figure 23. Transactions by area and month over time (month) [68]. Per month, 2025-04-01 to 2026-09-01. Source: Transactions by area and month (432 features). The data is in the appendix.*

![Emerging hot spots: ATM faults](charts/03-emerging-hot-spots-atm-faults.png)

*Figure 24. Emerging hot spots: ATM faults [69]. Source: Emerging hot spots: ATM faults (84 features). The data is in the appendix.*

![Emerging hot spots: New customer accounts](charts/04-emerging-hot-spots-new-customer-accounts.png)

*Figure 25. Emerging hot spots: New customer accounts [70]. Source: Emerging hot spots: New customer accounts (106 features). The data is in the appendix.*

![Summarize within: Service areas](charts/05-summarize-within-service-areas.png)

*Figure 26. Summarize within: Service areas [71]. Source: Summarize within: Service areas (24 features). The data is in the appendix.*

![Summarize within: Service areas](charts/06-summarize-within-service-areas.png)

*Figure 27. Summarize within: Service areas [72]. Source: Summarize within: Service areas (24 features). The data is in the appendix.*

![Summarize within: Service areas](charts/07-summarize-within-service-areas.png)

*Figure 28. Summarize within: Service areas [73]. Source: Summarize within: Service areas (24 features). The data is in the appendix.*

![Hot spot analysis: ATM faults](charts/08-hot-spot-analysis-atm-faults.png)

*Figure 29. Hot spot analysis: ATM faults [74]. Source: Hot spot analysis: ATM faults (240 features). The data is in the appendix.*

![Hot spot analysis: ATMs](charts/09-hot-spot-analysis-atms.png)

*Figure 30. Hot spot analysis: ATMs [75]. Source: Hot spot analysis: ATMs (68 features). The data is in the appendix.*

![Hot spot analysis: Branches](charts/10-hot-spot-analysis-branches.png)

*Figure 31. Hot spot analysis: Branches [76]. Source: Hot spot analysis: Branches (16 features). The data is in the appendix.*

![Calculate density: ATM faults](charts/11-calculate-density-atm-faults.png)

*Figure 32. Calculate density: ATM faults [77]. Source: Calculate density: ATM faults (1,893 features). The data is in the appendix.*

![Calculate density: ATMs](charts/12-calculate-density-atms.png)

*Figure 33. Calculate density: ATMs [78]. Source: Calculate density: ATMs (459 features). The data is in the appendix.*

![Calculate density: Branches](charts/13-calculate-density-branches.png)

*Figure 34. Calculate density: Branches [79]. Source: Calculate density: Branches (88 features). The data is in the appendix.*

![Find point clusters: ATM faults](charts/14-find-point-clusters-atm-faults.png)

*Figure 35. Find point clusters: ATM faults [80]. Period: 2026-04-01 to 2026-09-30 (reported_at). Source: Find point clusters: ATM faults (650 features). The data is in the appendix.*

![Find point clusters: cluster hulls](charts/15-find-point-clusters-cluster-hulls.png)

*Figure 36. Find point clusters: cluster hulls [81]. Source: Find point clusters: cluster hulls (14 features). The data is in the appendix.*

![Find point clusters: ATMs](charts/16-find-point-clusters-atms.png)

*Figure 37. Find point clusters: ATMs [82]. Source: Find point clusters: ATMs (140 features). The data is in the appendix.*

![Find point clusters: cluster hulls](charts/17-find-point-clusters-cluster-hulls.png)

*Figure 38. Find point clusters: cluster hulls [83]. Source: Find point clusters: cluster hulls (1 features). The data is in the appendix.*

![Find point clusters: Branches](charts/18-find-point-clusters-branches.png)

*Figure 39. Find point clusters: Branches [84]. Source: Find point clusters: Branches (30 features). The data is in the appendix.*

![Find point clusters: cluster hulls](charts/19-find-point-clusters-cluster-hulls.png)

*Figure 40. Find point clusters: cluster hulls [85]. Source: Find point clusters: cluster hulls (1 features). The data is in the appendix.*

![Find outliers: Service areas](charts/20-find-outliers-service-areas.png)

*Figure 41. Find outliers: Service areas [86]. Source: Find outliers: Service areas (24 features). The data is in the appendix.*

![Repeat and near repeat classification: ATM faults](charts/21-repeat-and-near-repeat-classification-atm-faults.png)

*Figure 42. Repeat and near repeat classification: ATM faults [87]. Per week, 2026-03-30 to 2026-09-28. Source: Repeat and near repeat classification: ATM faults (650 features). The data is in the appendix.*

![Repeat and near repeat classification: ATM faults over time](charts/22-repeat-and-near-repeat-classification-atm-faults-over-time.png)

*Figure 43. Repeat and near repeat classification: ATM faults over time [88]. Per week, 2026-03-30 to 2026-09-28. Source: Repeat and near repeat classification: ATM faults (650 features). The data is in the appendix.*

![Find point clusters: ATM faults over time](charts/23-find-point-clusters-atm-faults-over-time.png)

*Figure 44. Find point clusters: ATM faults over time [89]. Per week, 2026-03-30 to 2026-09-28. Source: Find point clusters: ATM faults (650 features). The data is in the appendix.*

## Changes since the last run

This is the first report of this project.

## Sources

[1] Rendering: Chart: Timeline: events per month by dataset. 54 rows (id 5350075e-73c1-474d-950e-d360606711fc)  
[2] Dataset: New customer accounts. vector, 1971 rows (id ca1a5fee-d2f1-4e7a-96ff-a2ae025365b9)  
[3] Dataset: ATM faults. vector, 650 rows (id 1a89d4d2-5bd2-4af8-8992-58556152d450)  
[4] Dataset: Transactions by area and month. table, 432 rows (id d861cdce-91fa-4661-9b4c-084d3fdfa2af)  
[5] Catalog: Emerging hot spots (operation catalog, gp_space_time_hot_spots). Where hot spots are new, consecutive, intensifying, persistent, diminishing, sporadic, oscillating or historical over time. ArcGIS: Emerging Hot Spot Analysis, Create Space Time Cube. (id gp_space_time_hot_spots)  
[6] Plan: Space time hot spots of ATM faults (the Start plan's candidate c_11dd443868e3). ATM faults carry dates in 'reported_at': emerging hot spot analysis finds new, intensifying and fading hot spots per week. (id 8966c973-8846-4a37-90ad-665c80260ba6:c_11dd443868e3)  
[7] Tool result: gp_space_time_hot_spots, step 0. Emerging hot spots: 1 layer saved: Emerging hot spots: ATM faults (84 features). (id b6f5a2bb-6262-4265-bcee-f2cfddd03ddb:0)  
[8] Layer: ATM faults. 650 features (id 2bdf3635-4c99-4580-a937-de55490e43f5)  
[9] Plan: Space time hot spots of New customer accounts (the Start plan's candidate c_bb1839f07414). New customer accounts carry dates in 'opened_on': emerging hot spot analysis finds new, intensifying and fading hot spots per month. (id 8966c973-8846-4a37-90ad-665c80260ba6:c_bb1839f07414)  
[10] Tool result: gp_space_time_hot_spots, step 1. Emerging hot spots: 1 layer saved: Emerging hot spots: New customer accounts (106 features). (id b6f5a2bb-6262-4265-bcee-f2cfddd03ddb:1)  
[11] Layer: New customer accounts. 1,971 features (id e03b0fae-60ec-4079-913e-15d82b331701)  
[12] Catalog: Summarize within (operation catalog, gp_summarize_within). Count and aggregate the features inside each polygon (or bin), proportionally for lines and polygons. ArcGIS: Summarize Within, Aggregate Points. QGIS: native:countpointsinpolygon, native:sumlinelengths. (id gp_summarize_within)  
[13] Plan: ATM faults per feature of Service areas (the Start plan's candidate c_b3b30286944f). Count ATM faults inside each feature of Service areas by fault; the counts map as a choropleth (natural breaks, sequential colors). (id 8966c973-8846-4a37-90ad-665c80260ba6:c_b3b30286944f)  
[14] Tool result: gp_summarize_within, step 0. Summarize within: 1 layer saved: Summarize within: Service areas (24 features). (id 55a6123f-a37d-4fea-bc56-f3134e981a4d:0)  
[15] Layer: Service areas. 24 features (id 6344158f-5202-4018-adf5-fef914f7efc7)  
[16] Tool result: gp_summarize_within, step 1. Summarize within: 1 layer saved: Summarize within: Service areas (24 features). (id 55a6123f-a37d-4fea-bc56-f3134e981a4d:1)  
[17] Layer: ATMs. 140 features (id f9f05311-d0d9-491b-b60e-76d6074147ae)  
[18] Tool result: gp_summarize_within, step 2. Summarize within: 1 layer saved: Summarize within: Service areas (24 features). (id 55a6123f-a37d-4fea-bc56-f3134e981a4d:2)  
[19] Layer: Branches. 30 features (id 8f400be9-7e82-4433-96cb-438a2b6ed3d3)  
[20] Catalog: Hot spot analysis (operation catalog, gp_hot_spots). Getis-Ord Gi* hot and cold spots with z-scores, p-values and confidence bins. ArcGIS: Find Hot Spots, Hot Spot Analysis (Getis-Ord Gi*). QGIS: qgis:hotspot (plugin). (id gp_hot_spots)  
[21] Plan: Hot spots of ATM faults (the Start plan's candidate c_33af6577a5c2). 650 points: Getis-Ord Gi* tests whether the concentrations in ATM faults are statistically significant. (id 8966c973-8846-4a37-90ad-665c80260ba6:c_33af6577a5c2)  
[22] Tool result: gp_hot_spots, step 0. Hot spot analysis: 1 layer saved: Hot spot analysis: ATM faults (240 features). (id 2f94438c-e0a6-41b8-a152-e3151a6def70:0)  
[23] Plan: Hot spots of ATMs (the Start plan's candidate c_eacb2979e4db). 140 points: Getis-Ord Gi* tests whether the concentrations in ATMs are statistically significant. (id 8966c973-8846-4a37-90ad-665c80260ba6:c_eacb2979e4db)  
[24] Tool result: gp_hot_spots, step 1. Hot spot analysis: 1 layer saved: Hot spot analysis: ATMs (68 features). (id 2f94438c-e0a6-41b8-a152-e3151a6def70:1)  
[25] Plan: Hot spots of Branches (the Start plan's candidate c_b3e06b1ef787). 30 points: Getis-Ord Gi* tests whether the concentrations in Branches are statistically significant. (id 8966c973-8846-4a37-90ad-665c80260ba6:c_b3e06b1ef787)  
[26] Tool result: gp_hot_spots, step 2. Hot spot analysis: 1 layer saved: Hot spot analysis: Branches (16 features). (id 2f94438c-e0a6-41b8-a152-e3151a6def70:2)  
[27] Catalog: Calculate density (operation catalog, gp_calculate_density). Kernel density per square kilometer on a grid. ArcGIS: Calculate Density, Kernel Density. QGIS: native:heatmapkerneldensityestimation. (id gp_calculate_density)  
[28] Plan: Kernel density of ATM faults (the Start plan's candidate c_984c04544f87). 650 points in ATM faults: a density surface shows where they concentrate. (id 8966c973-8846-4a37-90ad-665c80260ba6:c_984c04544f87)  
[29] Tool result: gp_calculate_density, step 0. Calculate density: 1 layer saved: Calculate density: ATM faults (1893 features). (id 8c87dbcb-4a64-4e28-b285-e529fd8b7cc4:0)  
[30] Plan: Kernel density of ATMs (the Start plan's candidate c_0c2ac7adbfb7). 140 points in ATMs: a density surface shows where they concentrate. (id 8966c973-8846-4a37-90ad-665c80260ba6:c_0c2ac7adbfb7)  
[31] Tool result: gp_calculate_density, step 1. Calculate density: 1 layer saved: Calculate density: ATMs (459 features). (id 8c87dbcb-4a64-4e28-b285-e529fd8b7cc4:1)  
[32] Plan: Kernel density of Branches (the Start plan's candidate c_fbae95ca186a). 30 points in Branches: a density surface shows where they concentrate. (id 8966c973-8846-4a37-90ad-665c80260ba6:c_fbae95ca186a)  
[33] Tool result: gp_calculate_density, step 2. Calculate density: 1 layer saved: Calculate density: Branches (88 features). (id 8c87dbcb-4a64-4e28-b285-e529fd8b7cc4:2)  
[34] Catalog: Find point clusters (operation catalog, gp_find_point_clusters). Density-based clusters (DBSCAN, HDBSCAN or OPTICS, optionally space-time) with hulls and noise. ArcGIS: Find Point Clusters, Density-based Clustering. QGIS: native:dbscanclustering, native:stdbscanclustering. (id gp_find_point_clusters)  
[35] Plan: Cluster points of ATM faults (the Start plan's candidate c_7a7e6112a7c4). DBSCAN finds dense groups of ATM faults and separates noise. (id 8966c973-8846-4a37-90ad-665c80260ba6:c_7a7e6112a7c4)  
[36] Tool result: gp_find_point_clusters, step 0. Find point clusters: 2 layers saved: Find point clusters: ATM faults (650 features); Find point clusters: cluster hulls (14 features). (id ddb39481-1f8b-4e18-afe9-2b1c4db1b0ef:0)  
[37] Plan: Cluster points of ATMs (the Start plan's candidate c_0807abb77372). DBSCAN finds dense groups of ATMs and separates noise. (id 8966c973-8846-4a37-90ad-665c80260ba6:c_0807abb77372)  
[38] Tool result: gp_find_point_clusters, step 1. Find point clusters: 2 layers saved: Find point clusters: ATMs (140 features); Find point clusters: cluster hulls (1 feature). (id ddb39481-1f8b-4e18-afe9-2b1c4db1b0ef:1)  
[39] Plan: Cluster points of Branches (the Start plan's candidate c_5c9c54534fb2). DBSCAN finds dense groups of Branches and separates noise. (id 8966c973-8846-4a37-90ad-665c80260ba6:c_5c9c54534fb2)  
[40] Tool result: gp_find_point_clusters, step 2. Find point clusters: 2 layers saved: Find point clusters: Branches (30 features); Find point clusters: cluster hulls (1 feature). (id ddb39481-1f8b-4e18-afe9-2b1c4db1b0ef:2)  
[41] Catalog: Find outliers (operation catalog, gp_outliers). Anselin local Moran’s I: HH, LL clusters and HL, LH outliers. ArcGIS: Find Outliers, Cluster and Outlier Analysis (Anselin Local Moran’s I). (id gp_outliers)  
[42] Plan: Local morans i of Service areas (the Start plan's candidate c_82c4828c335b). Local Moran's I maps clusters and outliers of 'population'. (id 8966c973-8846-4a37-90ad-665c80260ba6:c_82c4828c335b)  
[43] Tool result: gp_outliers, step 0. Find outliers: 1 layer saved: Find outliers: Service areas (24 features). (id ebdb002d-2e5b-42ca-870b-a29352cb798d:0)  
[44] Catalog: Repeat and near repeat classification (operation catalog, gp_near_repeat). Classify incidents as originators, repeats (same place) or near repeats within a distance and time window, with a Knox test table. ArcGIS: Repeat and Near Repeat Classification, Near Repeat Calculator. (id gp_near_repeat)  
[45] Plan: Near repeat of ATM faults (the Start plan's candidate c_0525e54ffa77). Near repeat analysis tests whether ATM faults close in space also follow each other within 18 days. (id 8966c973-8846-4a37-90ad-665c80260ba6:c_0525e54ffa77)  
[46] Tool result: gp_near_repeat, step 0. Repeat and near repeat classification: 1 layer saved: Repeat and near repeat classification: ATM faults (650 features). (id decafdf2-8712-4857-a4f0-3ed34c0aec6c:0)  
[47] Rendering: Map: Emerging hot spots: ATM faults (id 76ee6f37-74c8-4c56-85ba-bc6f47108994)  
[48] Rendering: Map: Emerging hot spots: New customer accounts (id 09264e18-ba78-46fb-a2ce-7ab906f160d8)  
[49] Rendering: Map: Summarize within: Service areas (id ef256995-4467-474b-80ab-8a608ad3fd72)  
[50] Rendering: Map: Summarize within: Service areas (id 35df054b-4186-4e03-a63c-67a7f2baaa23)  
[51] Rendering: Map: Summarize within: Service areas (id 6fbc5d26-341e-436b-b24d-678ad5b8ef4b)  
[52] Rendering: Map: Service areas by population (id b341088d-2db3-4a29-8635-157f59123b60)  
[53] Rendering: Map: Hot spot analysis: ATM faults (id b6494a6c-b71f-443e-b39b-2d54712e5f62)  
[54] Rendering: Map: Hot spot analysis: ATMs (id ca6b7c1c-7a7b-4dc0-96b2-9562dc7d9dda)  
[55] Rendering: Map: Hot spot analysis: Branches (id 71bb9dc5-68ce-41c3-b0e3-97c35130725d)  
[56] Rendering: Map: Calculate density: ATM faults (id 61e8469a-ca23-498c-a844-56c1eb1808e4)  
[57] Rendering: Map: Calculate density: ATMs (id 050ba628-bb1f-487e-a378-542769c03674)  
[58] Rendering: Map: Calculate density: Branches (id 2bb0e1f9-4b39-400e-8731-95ab5d3eb007)  
[59] Rendering: Map: Find point clusters: ATM faults (id 4a3ddce9-b286-4982-b29f-e520074e1420)  
[60] Rendering: Map: Find point clusters: cluster hulls (id 60fbd3c3-41cf-4188-af84-ac3eb9ec45f4)  
[61] Rendering: Map: Find point clusters: ATMs (id f933545e-ee47-43ac-8a7c-7f5ab4f3a26a)  
[62] Rendering: Map: Find point clusters: cluster hulls (id 712b47f2-0a18-4783-bd5f-42b3ab911373)  
[63] Rendering: Map: Find point clusters: Branches (id baa9825c-0a0a-41a0-8deb-28d4fe342b5d)  
[64] Rendering: Map: Find point clusters: cluster hulls (id b7e4a16c-e3ce-489f-8e8e-7709e70f30c5)  
[65] Rendering: Map: Find outliers: Service areas (id 097cd92c-bdca-494e-8c39-c6c9f8f013ff)  
[66] Rendering: Map: Repeat and near repeat classification: ATM faults (id c50405e6-0141-4a62-a6fa-302fdbda1f8e)  
[67] Rendering: Chart: ATM faults over time (reported_at). 108 rows (id 7e58d05f-f91f-4b43-b9a7-53ede05aab3c)  
[68] Rendering: Chart: Transactions by area and month over time (month). 18 rows (id 43d54876-a680-44f0-8807-97d30add6c7b)  
[69] Rendering: Chart: Emerging hot spots: ATM faults. 8 rows (id 05240352-4094-459b-93b9-b0b39ff66388)  
[70] Rendering: Chart: Emerging hot spots: New customer accounts. 8 rows (id d171a240-e329-4555-b7f7-dd2e1a4eb630)  
[71] Rendering: Chart: Summarize within: Service areas. 6 rows (id 34d3fffd-c931-4253-8f1c-b9a3d02f33a2)  
[72] Rendering: Chart: Summarize within: Service areas. 6 rows (id cfdf43ab-0f4b-483a-b49f-f59ae44a4cb6)  
[73] Rendering: Chart: Summarize within: Service areas. 6 rows (id 102d8194-390f-4002-a602-f35d93571a68)  
[74] Rendering: Chart: Hot spot analysis: ATM faults. 16 rows (id afea17ba-ef2f-4c29-a788-e9eb93790b0a)  
[75] Rendering: Chart: Hot spot analysis: ATMs. 9 rows (id 06f70b45-9da8-41dd-b783-99cdc39a5ed1)  
[76] Rendering: Chart: Hot spot analysis: Branches. 5 rows (id 8dac2f16-e156-4c0c-a3bc-1abdd53acf19)  
[77] Rendering: Chart: Calculate density: ATM faults. 12 rows (id 5300cc79-ddc9-4789-ba0e-8117219edc08)  
[78] Rendering: Chart: Calculate density: ATMs. 10 rows (id 85a5ca6f-877a-4152-ad20-9962042d8773)  
[79] Rendering: Chart: Calculate density: Branches. 8 rows (id a4e1be4d-2170-4fa2-8eaf-e621a6e88889)  
[80] Rendering: Chart: Find point clusters: ATM faults. 15 rows (id 7475dab8-8ac2-451a-896a-2cba2b21ceb0)  
[81] Rendering: Chart: Find point clusters: cluster hulls. 14 rows (id 050ea633-9bb7-4f6f-b729-e48e19e2f29f)  
[82] Rendering: Chart: Find point clusters: ATMs. 1 rows (id 0a6689e7-beeb-407c-a93a-968bb7b5fd18)  
[83] Rendering: Chart: Find point clusters: cluster hulls. 1 rows (id b2c962a7-8efd-4811-b98b-b95c07a42f91)  
[84] Rendering: Chart: Find point clusters: Branches. 1 rows (id 51e26a99-7c7f-4f46-b42b-db4446790e9b)  
[85] Rendering: Chart: Find point clusters: cluster hulls. 1 rows (id 74c8188c-0eb1-4132-b48c-c6f932d42986)  
[86] Rendering: Chart: Find outliers: Service areas. 24 rows (id 685cf60b-64b0-4c10-abc0-8d35b5d3fe39)  
[87] Rendering: Chart: Repeat and near repeat classification: ATM faults. 108 rows (id 5fedd8da-8a3c-43d9-b089-3ef291f6037b)  
[88] Rendering: Chart: Repeat and near repeat classification: ATM faults over time. 108 rows (id 39e30229-db4d-4477-9cf4-6a0e357a9214)  
[89] Rendering: Chart: Find point clusters: ATM faults over time. 108 rows (id 15d42b9c-5969-49e7-92a7-e2d8d7ee3bed)  
[90] Dataset: Branches. vector, 30 rows (id 723808dc-deb7-4f84-8a71-4d9971c7481d)  
[91] Dataset: ATMs. vector, 140 rows (id 67e0988a-c269-4f42-9feb-d3f8cc8e6cce)  
[92] Dataset: Service areas. vector, 24 rows (id edb72b32-4d2f-4324-82c3-07d95a35855a)  
[93] Dataset: Business clients. vector, 259 rows (id fc7b0810-e2b6-4b63-b451-6145da616183)  
[94] Dataset: Competitor branches. vector, 80 rows (id dc85622c-0f83-43c4-b78a-de81f9514dc0)  
[95] Dataset: Branch targets. table, 90 rows (id 1805a967-c33f-4bec-bae7-33ddb672f5c2)  
[96] Dataset: Suburban rail lines. vector, 3 rows (id be0b9524-ec6b-46e9-8d03-12477e584565)  
[97] Dataset: Branch network strategy memo. document, ? rows (id a09d49c1-003f-46cd-89de-81aa969e1f47)  
[98] Record: Sample: Banking, Mumbai (id 36cc8b7b-aabc-4b71-981b-a2c53e8f31ad)  
[99] Layer: Emerging hot spots: ATM faults. 84 features (id 7a9e76b7-6a9c-4142-9b1b-2ba16950826e)  
[100] Layer: Emerging hot spots: New customer accounts. 106 features (id d2737b97-6e34-4724-9cc1-4fc26dc27ef0)  
[101] Layer: Summarize within: Service areas. 24 features (id 82ed72b3-93b8-44f3-aed9-b04c683016ac)  
[102] Layer: Summarize within: Service areas. 24 features (id 65d1f682-3dbc-430d-9611-73e246f61c59)  
[103] Layer: Summarize within: Service areas. 24 features (id b0085626-4a10-4142-a439-fee139d0d654)  
[104] Layer: Hot spot analysis: ATM faults. 240 features (id 2f93886f-42ab-46e3-9a7f-1af360604373)  
[105] Layer: Hot spot analysis: ATMs. 68 features (id a1c55487-2858-4142-ad20-bd6b5d7310eb)  
[106] Layer: Hot spot analysis: Branches. 16 features (id 35e39240-1bef-45a6-a63c-a0d865426146)  
[107] Layer: Calculate density: ATM faults. 1,893 features (id 8c0ee5b1-c9d1-478d-b5b1-c98fa86bd9f8)  
[108] Layer: Calculate density: ATMs. 459 features (id f79c3fc8-1034-4abe-a828-0eb4004201cd)  
[109] Layer: Calculate density: Branches. 88 features (id 1902ddf0-4227-4a92-ac7c-939bc604a5b3)  
[110] Layer: Find point clusters: ATM faults. 650 features (id 05658f04-f9ea-4c40-a4c5-30b7e4e77625)  
[111] Layer: Find point clusters: cluster hulls. 14 features (id de772841-b4f1-4f69-8eba-cfe862a45c01)  
[112] Layer: Find point clusters: ATMs. 140 features (id 4db94fe3-4761-4f5a-9ec9-577c48d7d13a)  
[113] Layer: Find point clusters: cluster hulls. 1 features (id a0af9a60-4b16-4861-a4d6-0aac7ae08c50)  
[114] Layer: Find point clusters: Branches. 30 features (id 139f426e-3703-4fb2-818e-11512ded6292)  
[115] Layer: Find point clusters: cluster hulls. 1 features (id 8dc3f93c-e454-4aa6-a12a-36f3ccd4a02a)  
[116] Layer: Find outliers: Service areas. 24 features (id 02024e38-4d51-481d-a141-856333b7c347)  
[117] Layer: Repeat and near repeat classification: ATM faults. 650 features (id c209894e-f1ec-4882-a054-494986b09a69)  

## Appendix: the data behind the charts

### Figure 1. Timeline: events per month by dataset [1]

| period | series | value |
| --- | --- | --- |
| 2025-04-01 | New customer accounts | 93 |
| 2025-04-01 | ATM faults | 0 |
| 2025-04-01 | Transactions by area and month | 24 |
| 2025-05-01 | New customer accounts | 89 |
| 2025-05-01 | ATM faults | 0 |
| 2025-05-01 | Transactions by area and month | 24 |
| 2025-06-01 | New customer accounts | 98 |
| 2025-06-01 | ATM faults | 0 |
| 2025-06-01 | Transactions by area and month | 24 |
| 2025-07-01 | New customer accounts | 97 |
| 2025-07-01 | ATM faults | 0 |
| 2025-07-01 | Transactions by area and month | 24 |
| 2025-08-01 | New customer accounts | 78 |
| 2025-08-01 | ATM faults | 0 |
| 2025-08-01 | Transactions by area and month | 24 |
| 2025-09-01 | New customer accounts | 106 |
| 2025-09-01 | ATM faults | 0 |
| 2025-09-01 | Transactions by area and month | 24 |
| 2025-10-01 | New customer accounts | 109 |
| 2025-10-01 | ATM faults | 0 |
| 2025-10-01 | Transactions by area and month | 24 |
| 2025-11-01 | New customer accounts | 105 |
| 2025-11-01 | ATM faults | 0 |
| 2025-11-01 | Transactions by area and month | 24 |
| 2025-12-01 | New customer accounts | 111 |
| 2025-12-01 | ATM faults | 0 |
| 2025-12-01 | Transactions by area and month | 24 |
| 2026-01-01 | New customer accounts | 114 |
| 2026-01-01 | ATM faults | 0 |
| 2026-01-01 | Transactions by area and month | 24 |
| 2026-02-01 | New customer accounts | 88 |
| 2026-02-01 | ATM faults | 0 |
| 2026-02-01 | Transactions by area and month | 24 |
| 2026-03-01 | New customer accounts | 106 |
| 2026-03-01 | ATM faults | 0 |
| 2026-03-01 | Transactions by area and month | 24 |
| 2026-04-01 | New customer accounts | 133 |
| 2026-04-01 | ATM faults | 101 |
| 2026-04-01 | Transactions by area and month | 24 |
| 2026-05-01 | New customer accounts | 118 |
| 2026-05-01 | ATM faults | 115 |
| 2026-05-01 | Transactions by area and month | 24 |
| 2026-06-01 | New customer accounts | 121 |
| 2026-06-01 | ATM faults | 100 |
| 2026-06-01 | Transactions by area and month | 24 |
| 2026-07-01 | New customer accounts | 130 |
| 2026-07-01 | ATM faults | 116 |
| 2026-07-01 | Transactions by area and month | 24 |
| 2026-08-01 | New customer accounts | 150 |
| 2026-08-01 | ATM faults | 119 |
| 2026-08-01 | Transactions by area and month | 24 |
| 2026-09-01 | New customer accounts | 125 |
| 2026-09-01 | ATM faults | 99 |
| 2026-09-01 | Transactions by area and month | 24 |

### Figure 22. ATM faults over time (reported_at) [67]

| period | series | value |
| --- | --- | --- |
| 2026-03-30 | cash out | 6 |
| 2026-03-30 | network | 4 |
| 2026-03-30 | card reader | 3 |
| 2026-03-30 | dispenser jam | 4 |
| 2026-04-06 | cash out | 6 |
| 2026-04-06 | network | 6 |
| 2026-04-06 | card reader | 8 |
| 2026-04-06 | dispenser jam | 12 |
| 2026-04-13 | cash out | 6 |
| 2026-04-13 | network | 5 |
| 2026-04-13 | card reader | 5 |
| 2026-04-13 | dispenser jam | 2 |
| 2026-04-20 | cash out | 6 |
| 2026-04-20 | network | 4 |
| 2026-04-20 | card reader | 4 |
| 2026-04-20 | dispenser jam | 4 |
| 2026-04-27 | cash out | 8 |
| 2026-04-27 | network | 6 |
| 2026-04-27 | card reader | 8 |
| 2026-04-27 | dispenser jam | 6 |
| 2026-05-04 | cash out | 6 |
| 2026-05-04 | network | 7 |
| 2026-05-04 | card reader | 6 |
| 2026-05-04 | dispenser jam | 4 |
| 2026-05-11 | cash out | 5 |
| 2026-05-11 | network | 8 |
| 2026-05-11 | card reader | 4 |
| 2026-05-11 | dispenser jam | 10 |
| 2026-05-18 | cash out | 7 |
| 2026-05-18 | network | 9 |
| 2026-05-18 | card reader | 7 |
| 2026-05-18 | dispenser jam | 6 |
| 2026-05-25 | cash out | 11 |
| 2026-05-25 | network | 6 |
| 2026-05-25 | card reader | 3 |
| 2026-05-25 | dispenser jam | 4 |
| 2026-06-01 | cash out | 6 |
| 2026-06-01 | network | 7 |
| 2026-06-01 | card reader | 7 |
| 2026-06-01 | dispenser jam | 2 |
| 2026-06-08 | cash out | 4 |
| 2026-06-08 | network | 6 |
| 2026-06-08 | card reader | 9 |
| 2026-06-08 | dispenser jam | 4 |
| 2026-06-15 | cash out | 5 |
| 2026-06-15 | network | 4 |
| 2026-06-15 | card reader | 6 |
| 2026-06-15 | dispenser jam | 6 |
| 2026-06-22 | cash out | 4 |
| 2026-06-22 | network | 10 |
| 2026-06-22 | card reader | 7 |
| 2026-06-22 | dispenser jam | 9 |
| 2026-06-29 | cash out | 7 |
| 2026-06-29 | network | 4 |
| 2026-06-29 | card reader | 9 |
| 2026-06-29 | dispenser jam | 4 |
| 2026-07-06 | cash out | 10 |
| 2026-07-06 | network | 7 |
| 2026-07-06 | card reader | 7 |
| 2026-07-06 | dispenser jam | 4 |
| 2026-07-13 | cash out | 8 |
| 2026-07-13 | network | 11 |
| 2026-07-13 | card reader | 6 |
| 2026-07-13 | dispenser jam | 6 |
| 2026-07-20 | cash out | 2 |
| 2026-07-20 | network | 6 |
| 2026-07-20 | card reader | 3 |
| 2026-07-20 | dispenser jam | 8 |
| 2026-07-27 | cash out | 5 |
| 2026-07-27 | network | 7 |
| 2026-07-27 | card reader | 8 |
| 2026-07-27 | dispenser jam | 5 |
| 2026-08-03 | cash out | 7 |
| 2026-08-03 | network | 5 |
| 2026-08-03 | card reader | 6 |
| 2026-08-03 | dispenser jam | 2 |
| 2026-08-10 | cash out | 8 |
| 2026-08-10 | network | 4 |
| 2026-08-10 | card reader | 7 |
| 2026-08-10 | dispenser jam | 8 |
| 2026-08-17 | cash out | 9 |
| 2026-08-17 | network | 8 |
| 2026-08-17 | card reader | 7 |
| 2026-08-17 | dispenser jam | 7 |
| 2026-08-24 | cash out | 9 |
| 2026-08-24 | network | 7 |
| 2026-08-24 | card reader | 8 |
| 2026-08-24 | dispenser jam | 7 |
| 2026-08-31 | cash out | 3 |
| 2026-08-31 | network | 7 |
| 2026-08-31 | card reader | 3 |
| 2026-08-31 | dispenser jam | 6 |
| 2026-09-07 | cash out | 7 |
| 2026-09-07 | network | 4 |
| 2026-09-07 | card reader | 2 |
| 2026-09-07 | dispenser jam | 5 |
| 2026-09-14 | cash out | 11 |
| 2026-09-14 | network | 13 |
| 2026-09-14 | card reader | 4 |
| 2026-09-14 | dispenser jam | 4 |
| 2026-09-21 | cash out | 4 |
| 2026-09-21 | network | 4 |
| 2026-09-21 | card reader | 9 |
| 2026-09-21 | dispenser jam | 5 |
| 2026-09-28 | cash out | 3 |
| 2026-09-28 | network | 3 |
| 2026-09-28 | card reader | 4 |
| 2026-09-28 | dispenser jam | 1 |

### Figure 23. Transactions by area and month over time (month) [68]

| period | value |
| --- | --- |
| 2025-04-01 | 3,954 |
| 2025-05-01 | 4,028 |
| 2025-06-01 | 4,107 |
| 2025-07-01 | 4,189 |
| 2025-08-01 | 4,267 |
| 2025-09-01 | 4,351 |
| 2025-10-01 | 4,438 |
| 2025-11-01 | 4,525 |
| 2025-12-01 | 4,618 |
| 2026-01-01 | 4,713 |
| 2026-02-01 | 4,809 |
| 2026-03-01 | 4,909 |
| 2026-04-01 | 5,012 |
| 2026-05-01 | 5,119 |
| 2026-06-01 | 5,230 |
| 2026-07-01 | 5,341 |
| 2026-08-01 | 5,458 |
| 2026-09-01 | 5,578 |

### Figure 24. Emerging hot spots: ATM faults [69]

| bin_start | bin_end | count |
| --- | --- | --- |
| -0.984 | -0.6045 | 45 |
| -0.6045 | -0.225 | 0 |
| -0.225 | 0.1545 | 16 |
| 0.1545 | 0.534 | 0 |
| 0.534 | 0.9135 | 15 |
| 0.9135 | 1.293 | 0 |
| 1.293 | 1.673 | 4 |
| 1.673 | 2.052 | 4 |

### Figure 25. Emerging hot spots: New customer accounts [70]

| bin_start | bin_end | count |
| --- | --- | --- |
| -1.425 | 0.1437 | 55 |
| 0.1437 | 1.712 | 34 |
| 1.712 | 3.281 | 4 |
| 3.281 | 4.85 | 6 |
| 4.85 | 6.419 | 1 |
| 6.419 | 7.987 | 2 |
| 7.987 | 9.556 | 1 |
| 9.556 | 11.12 | 3 |

### Figure 26. Summarize within: Service areas [71]

| bin_start | bin_end | count |
| --- | --- | --- |
| 0 | 13.83 | 11 |
| 13.83 | 27.67 | 5 |
| 27.67 | 41.5 | 2 |
| 41.5 | 55.33 | 1 |
| 55.33 | 69.17 | 2 |
| 69.17 | 83 | 3 |

### Figure 27. Summarize within: Service areas [72]

| bin_start | bin_end | count |
| --- | --- | --- |
| 0 | 1.833 | 2 |
| 1.833 | 3.667 | 5 |
| 3.667 | 5.5 | 3 |
| 5.5 | 7.333 | 6 |
| 7.333 | 9.167 | 4 |
| 9.167 | 11 | 4 |

### Figure 28. Summarize within: Service areas [73]

| bin_start | bin_end | count |
| --- | --- | --- |
| 0 | 0.6667 | 6 |
| 0.6667 | 1.333 | 10 |
| 1.333 | 2 | 0 |
| 2 | 2.667 | 5 |
| 2.667 | 3.333 | 2 |
| 3.333 | 4 | 1 |

### Figure 29. Hot spot analysis: ATM faults [74]

| bin_start | bin_end | band | band_label | count |
| --- | --- | --- | --- | --- |
| -2 | -1.5 | cold_90 | Cold spot, 90% confidence | 3 |
| -2 | -1.5 | not_significant | Not significant | 5 |
| -1.5 | -1 | not_significant | Not significant | 59 |
| -1 | -0.5 | not_significant | Not significant | 56 |
| -0.5 | 0 | not_significant | Not significant | 19 |
| 0 | 0.5 | not_significant | Not significant | 12 |
| 0.5 | 1 | not_significant | Not significant | 18 |
| 1 | 1.5 | not_significant | Not significant | 18 |
| 1.5 | 2 | hot_90 | Hot spot, 90% confidence | 13 |
| 1.5 | 2 | hot_95 | Hot spot, 95% confidence | 1 |
| 1.5 | 2 | not_significant | Not significant | 8 |
| 2 | 2.5 | hot_95 | Hot spot, 95% confidence | 15 |
| 2.5 | 3 | hot_95 | Hot spot, 95% confidence | 2 |
| 2.5 | 3 | hot_99 | Hot spot, 99% confidence | 5 |
| 3 | 3.5 | hot_99 | Hot spot, 99% confidence | 5 |
| 4 | 4.5 | hot_99 | Hot spot, 99% confidence | 1 |

### Figure 30. Hot spot analysis: ATMs [75]

| bin_start | bin_end | band | band_label | count |
| --- | --- | --- | --- | --- |
| -2 | -1.5 | not_significant | Not significant | 1 |
| -1.5 | -1 | not_significant | Not significant | 3 |
| -1 | -0.5 | not_significant | Not significant | 10 |
| -0.5 | 0 | not_significant | Not significant | 10 |
| 0 | 0.5 | not_significant | Not significant | 18 |
| 0.5 | 1 | not_significant | Not significant | 16 |
| 1 | 1.5 | not_significant | Not significant | 8 |
| 1.5 | 2 | hot_90 | Hot spot, 90% confidence | 1 |
| 2 | 2.5 | hot_95 | Hot spot, 95% confidence | 1 |

### Figure 31. Hot spot analysis: Branches [76]

| bin_start | bin_end | band | band_label | count |
| --- | --- | --- | --- | --- |
| -1.5 | -1 | not_significant | Not significant | 4 |
| -1 | -0.5 | not_significant | Not significant | 3 |
| 0 | 0.5 | not_significant | Not significant | 4 |
| 0.5 | 1 | not_significant | Not significant | 1 |
| 1 | 1.5 | not_significant | Not significant | 4 |

### Figure 32. Calculate density: ATM faults [77]

| bin_start | bin_end | count |
| --- | --- | --- |
| 0 | 3.917 | 91 |
| 3.917 | 7.833 | 151 |
| 7.833 | 11.75 | 167 |
| 11.75 | 15.67 | 177 |
| 15.67 | 19.58 | 167 |
| 19.58 | 23.5 | 177 |
| 23.5 | 27.42 | 165 |
| 27.42 | 31.33 | 178 |
| 31.33 | 35.25 | 189 |
| 35.25 | 39.17 | 184 |
| 39.17 | 43.08 | 167 |
| 43.08 | 47 | 80 |

### Figure 33. Calculate density: ATMs [78]

| bin_start | bin_end | count |
| --- | --- | --- |
| 0 | 2.2 | 43 |
| 2.2 | 4.4 | 40 |
| 4.4 | 6.6 | 44 |
| 6.6 | 8.8 | 43 |
| 8.8 | 11 | 43 |
| 11 | 13.2 | 66 |
| 13.2 | 15.4 | 45 |
| 15.4 | 17.6 | 46 |
| 17.6 | 19.8 | 42 |
| 19.8 | 22 | 47 |

### Figure 34. Calculate density: Branches [79]

| bin_start | bin_end | count |
| --- | --- | --- |
| 1.546e-06 | 0.1222 | 27 |
| 0.1222 | 0.2445 | 15 |
| 0.2445 | 0.3667 | 11 |
| 0.3667 | 0.4889 | 6 |
| 0.4889 | 0.6111 | 10 |
| 0.6111 | 0.7334 | 5 |
| 0.7334 | 0.8556 | 11 |
| 0.8556 | 0.9778 | 3 |

### Figure 35. Find point clusters: ATM faults [80]

| cluster | features | fraction | share |
| --- | --- | --- | --- |
| Noise | 232 | 0.3569 | 35.7% |
| 5 | 52 | 0.08 | 8.0% |
| 7 | 42 | 0.06462 | 6.5% |
| 9 | 39 | 0.06 | 6.0% |
| 0 | 36 | 0.05538 | 5.5% |
| 1 | 36 | 0.05538 | 5.5% |
| 8 | 34 | 0.05231 | 5.2% |
| 11 | 28 | 0.04308 | 4.3% |
| 3 | 28 | 0.04308 | 4.3% |
| 4 | 28 | 0.04308 | 4.3% |
| 6 | 23 | 0.03538 | 3.5% |
| 12 | 20 | 0.03077 | 3.1% |
| 13 | 18 | 0.02769 | 2.8% |
| 10 | 17 | 0.02615 | 2.6% |
| 2 | 17 | 0.02615 | 2.6% |

### Figure 36. Find point clusters: cluster hulls [81]

| cluster | features | fraction | share |
| --- | --- | --- | --- |
| 0 | 1 | 0.07143 | 7.1% |
| 1 | 1 | 0.07143 | 7.1% |
| 10 | 1 | 0.07143 | 7.1% |
| 11 | 1 | 0.07143 | 7.1% |
| 12 | 1 | 0.07143 | 7.1% |
| 13 | 1 | 0.07143 | 7.1% |
| 2 | 1 | 0.07143 | 7.1% |
| 3 | 1 | 0.07143 | 7.1% |
| 4 | 1 | 0.07143 | 7.1% |
| 5 | 1 | 0.07143 | 7.1% |
| 6 | 1 | 0.07143 | 7.1% |
| 7 | 1 | 0.07143 | 7.1% |
| 8 | 1 | 0.07143 | 7.1% |
| 9 | 1 | 0.07143 | 7.1% |

### Figure 37. Find point clusters: ATMs [82]

| cluster | features | fraction | share |
| --- | --- | --- | --- |
| 0 | 140 | 1 | 100.0% |

### Figure 38. Find point clusters: cluster hulls [83]

| cluster | features | fraction | share |
| --- | --- | --- | --- |
| 0 | 1 | 1 | 100.0% |

### Figure 39. Find point clusters: Branches [84]

| cluster | features | fraction | share |
| --- | --- | --- | --- |
| 0 | 30 | 1 | 100.0% |

### Figure 40. Find point clusters: cluster hulls [85]

| cluster | features | fraction | share |
| --- | --- | --- | --- |
| 0 | 1 | 1 | 100.0% |

### Figure 41. Find outliers: Service areas [86]

| value | lag | type |
| --- | --- | --- |
| 0.7175 | 0.2147 | NS |
| -0.7605 | 0.6028 | NS |
| 0.3397 | 1.256 | NS |
| -0.0948 | 0.1267 | NS |
| 0.3969 | 0.5444 | NS |
| 1.671 | -0.0215 | NS |
| -0.1173 | 0.2867 | NS |
| -0.3478 | -0.0641 | NS |
| -0.7889 | -0.9745 | NS |
| -1.23 | -0.3683 | NS |
| -0.6027 | -0.6988 | NS |
| 0.6917 | 0.8524 | NS |
| 0.3711 | -0.7605 | NS |
| 0.6284 | -0.2384 | NS |
| 1.821 | 0.5157 | NS |
| -1.793 | -0.7222 | NS |
| -1.091 | -0.9646 | NS |
| 2.097 | -0.0948 | NS |
| -0.6233 | -0.5452 | NS |
| -0.8751 | -0.8528 | NS |
| -0.8417 | -0.5287 | NS |
| 1.031 | -0.4806 | NS |
| 0.3971 | 0.3969 | NS |
| -0.9956 | -0.5311 | NS |

### Figure 42. Repeat and near repeat classification: ATM faults [87]

| period | series | value |
| --- | --- | --- |
| 2026-03-30 | cash out | 6 |
| 2026-03-30 | network | 4 |
| 2026-03-30 | card reader | 3 |
| 2026-03-30 | dispenser jam | 4 |
| 2026-04-06 | cash out | 6 |
| 2026-04-06 | network | 6 |
| 2026-04-06 | card reader | 8 |
| 2026-04-06 | dispenser jam | 12 |
| 2026-04-13 | cash out | 6 |
| 2026-04-13 | network | 5 |
| 2026-04-13 | card reader | 5 |
| 2026-04-13 | dispenser jam | 2 |
| 2026-04-20 | cash out | 6 |
| 2026-04-20 | network | 4 |
| 2026-04-20 | card reader | 4 |
| 2026-04-20 | dispenser jam | 4 |
| 2026-04-27 | cash out | 8 |
| 2026-04-27 | network | 6 |
| 2026-04-27 | card reader | 8 |
| 2026-04-27 | dispenser jam | 6 |
| 2026-05-04 | cash out | 6 |
| 2026-05-04 | network | 7 |
| 2026-05-04 | card reader | 6 |
| 2026-05-04 | dispenser jam | 4 |
| 2026-05-11 | cash out | 5 |
| 2026-05-11 | network | 8 |
| 2026-05-11 | card reader | 4 |
| 2026-05-11 | dispenser jam | 10 |
| 2026-05-18 | cash out | 7 |
| 2026-05-18 | network | 9 |
| 2026-05-18 | card reader | 7 |
| 2026-05-18 | dispenser jam | 6 |
| 2026-05-25 | cash out | 11 |
| 2026-05-25 | network | 6 |
| 2026-05-25 | card reader | 3 |
| 2026-05-25 | dispenser jam | 4 |
| 2026-06-01 | cash out | 6 |
| 2026-06-01 | network | 7 |
| 2026-06-01 | card reader | 7 |
| 2026-06-01 | dispenser jam | 2 |
| 2026-06-08 | cash out | 4 |
| 2026-06-08 | network | 6 |
| 2026-06-08 | card reader | 9 |
| 2026-06-08 | dispenser jam | 4 |
| 2026-06-15 | cash out | 5 |
| 2026-06-15 | network | 4 |
| 2026-06-15 | card reader | 6 |
| 2026-06-15 | dispenser jam | 6 |
| 2026-06-22 | cash out | 4 |
| 2026-06-22 | network | 10 |
| 2026-06-22 | card reader | 7 |
| 2026-06-22 | dispenser jam | 9 |
| 2026-06-29 | cash out | 7 |
| 2026-06-29 | network | 4 |
| 2026-06-29 | card reader | 9 |
| 2026-06-29 | dispenser jam | 4 |
| 2026-07-06 | cash out | 10 |
| 2026-07-06 | network | 7 |
| 2026-07-06 | card reader | 7 |
| 2026-07-06 | dispenser jam | 4 |
| 2026-07-13 | cash out | 8 |
| 2026-07-13 | network | 11 |
| 2026-07-13 | card reader | 6 |
| 2026-07-13 | dispenser jam | 6 |
| 2026-07-20 | cash out | 2 |
| 2026-07-20 | network | 6 |
| 2026-07-20 | card reader | 3 |
| 2026-07-20 | dispenser jam | 8 |
| 2026-07-27 | cash out | 5 |
| 2026-07-27 | network | 7 |
| 2026-07-27 | card reader | 8 |
| 2026-07-27 | dispenser jam | 5 |
| 2026-08-03 | cash out | 7 |
| 2026-08-03 | network | 5 |
| 2026-08-03 | card reader | 6 |
| 2026-08-03 | dispenser jam | 2 |
| 2026-08-10 | cash out | 8 |
| 2026-08-10 | network | 4 |
| 2026-08-10 | card reader | 7 |
| 2026-08-10 | dispenser jam | 8 |
| 2026-08-17 | cash out | 9 |
| 2026-08-17 | network | 8 |
| 2026-08-17 | card reader | 7 |
| 2026-08-17 | dispenser jam | 7 |
| 2026-08-24 | cash out | 9 |
| 2026-08-24 | network | 7 |
| 2026-08-24 | card reader | 8 |
| 2026-08-24 | dispenser jam | 7 |
| 2026-08-31 | cash out | 3 |
| 2026-08-31 | network | 7 |
| 2026-08-31 | card reader | 3 |
| 2026-08-31 | dispenser jam | 6 |
| 2026-09-07 | cash out | 7 |
| 2026-09-07 | network | 4 |
| 2026-09-07 | card reader | 2 |
| 2026-09-07 | dispenser jam | 5 |
| 2026-09-14 | cash out | 11 |
| 2026-09-14 | network | 13 |
| 2026-09-14 | card reader | 4 |
| 2026-09-14 | dispenser jam | 4 |
| 2026-09-21 | cash out | 4 |
| 2026-09-21 | network | 4 |
| 2026-09-21 | card reader | 9 |
| 2026-09-21 | dispenser jam | 5 |
| 2026-09-28 | cash out | 3 |
| 2026-09-28 | network | 3 |
| 2026-09-28 | card reader | 4 |
| 2026-09-28 | dispenser jam | 1 |

### Figure 43. Repeat and near repeat classification: ATM faults over time [88]

| period | series | value |
| --- | --- | --- |
| 2026-03-30 | cash out | 6 |
| 2026-03-30 | network | 4 |
| 2026-03-30 | card reader | 3 |
| 2026-03-30 | dispenser jam | 4 |
| 2026-04-06 | cash out | 6 |
| 2026-04-06 | network | 6 |
| 2026-04-06 | card reader | 8 |
| 2026-04-06 | dispenser jam | 12 |
| 2026-04-13 | cash out | 6 |
| 2026-04-13 | network | 5 |
| 2026-04-13 | card reader | 5 |
| 2026-04-13 | dispenser jam | 2 |
| 2026-04-20 | cash out | 6 |
| 2026-04-20 | network | 4 |
| 2026-04-20 | card reader | 4 |
| 2026-04-20 | dispenser jam | 4 |
| 2026-04-27 | cash out | 8 |
| 2026-04-27 | network | 6 |
| 2026-04-27 | card reader | 8 |
| 2026-04-27 | dispenser jam | 6 |
| 2026-05-04 | cash out | 6 |
| 2026-05-04 | network | 7 |
| 2026-05-04 | card reader | 6 |
| 2026-05-04 | dispenser jam | 4 |
| 2026-05-11 | cash out | 5 |
| 2026-05-11 | network | 8 |
| 2026-05-11 | card reader | 4 |
| 2026-05-11 | dispenser jam | 10 |
| 2026-05-18 | cash out | 7 |
| 2026-05-18 | network | 9 |
| 2026-05-18 | card reader | 7 |
| 2026-05-18 | dispenser jam | 6 |
| 2026-05-25 | cash out | 11 |
| 2026-05-25 | network | 6 |
| 2026-05-25 | card reader | 3 |
| 2026-05-25 | dispenser jam | 4 |
| 2026-06-01 | cash out | 6 |
| 2026-06-01 | network | 7 |
| 2026-06-01 | card reader | 7 |
| 2026-06-01 | dispenser jam | 2 |
| 2026-06-08 | cash out | 4 |
| 2026-06-08 | network | 6 |
| 2026-06-08 | card reader | 9 |
| 2026-06-08 | dispenser jam | 4 |
| 2026-06-15 | cash out | 5 |
| 2026-06-15 | network | 4 |
| 2026-06-15 | card reader | 6 |
| 2026-06-15 | dispenser jam | 6 |
| 2026-06-22 | cash out | 4 |
| 2026-06-22 | network | 10 |
| 2026-06-22 | card reader | 7 |
| 2026-06-22 | dispenser jam | 9 |
| 2026-06-29 | cash out | 7 |
| 2026-06-29 | network | 4 |
| 2026-06-29 | card reader | 9 |
| 2026-06-29 | dispenser jam | 4 |
| 2026-07-06 | cash out | 10 |
| 2026-07-06 | network | 7 |
| 2026-07-06 | card reader | 7 |
| 2026-07-06 | dispenser jam | 4 |
| 2026-07-13 | cash out | 8 |
| 2026-07-13 | network | 11 |
| 2026-07-13 | card reader | 6 |
| 2026-07-13 | dispenser jam | 6 |
| 2026-07-20 | cash out | 2 |
| 2026-07-20 | network | 6 |
| 2026-07-20 | card reader | 3 |
| 2026-07-20 | dispenser jam | 8 |
| 2026-07-27 | cash out | 5 |
| 2026-07-27 | network | 7 |
| 2026-07-27 | card reader | 8 |
| 2026-07-27 | dispenser jam | 5 |
| 2026-08-03 | cash out | 7 |
| 2026-08-03 | network | 5 |
| 2026-08-03 | card reader | 6 |
| 2026-08-03 | dispenser jam | 2 |
| 2026-08-10 | cash out | 8 |
| 2026-08-10 | network | 4 |
| 2026-08-10 | card reader | 7 |
| 2026-08-10 | dispenser jam | 8 |
| 2026-08-17 | cash out | 9 |
| 2026-08-17 | network | 8 |
| 2026-08-17 | card reader | 7 |
| 2026-08-17 | dispenser jam | 7 |
| 2026-08-24 | cash out | 9 |
| 2026-08-24 | network | 7 |
| 2026-08-24 | card reader | 8 |
| 2026-08-24 | dispenser jam | 7 |
| 2026-08-31 | cash out | 3 |
| 2026-08-31 | network | 7 |
| 2026-08-31 | card reader | 3 |
| 2026-08-31 | dispenser jam | 6 |
| 2026-09-07 | cash out | 7 |
| 2026-09-07 | network | 4 |
| 2026-09-07 | card reader | 2 |
| 2026-09-07 | dispenser jam | 5 |
| 2026-09-14 | cash out | 11 |
| 2026-09-14 | network | 13 |
| 2026-09-14 | card reader | 4 |
| 2026-09-14 | dispenser jam | 4 |
| 2026-09-21 | cash out | 4 |
| 2026-09-21 | network | 4 |
| 2026-09-21 | card reader | 9 |
| 2026-09-21 | dispenser jam | 5 |
| 2026-09-28 | cash out | 3 |
| 2026-09-28 | network | 3 |
| 2026-09-28 | card reader | 4 |
| 2026-09-28 | dispenser jam | 1 |

### Figure 44. Find point clusters: ATM faults over time [89]

| period | series | value |
| --- | --- | --- |
| 2026-03-30 | cash out | 6 |
| 2026-03-30 | network | 4 |
| 2026-03-30 | card reader | 3 |
| 2026-03-30 | dispenser jam | 4 |
| 2026-04-06 | cash out | 6 |
| 2026-04-06 | network | 6 |
| 2026-04-06 | card reader | 8 |
| 2026-04-06 | dispenser jam | 12 |
| 2026-04-13 | cash out | 6 |
| 2026-04-13 | network | 5 |
| 2026-04-13 | card reader | 5 |
| 2026-04-13 | dispenser jam | 2 |
| 2026-04-20 | cash out | 6 |
| 2026-04-20 | network | 4 |
| 2026-04-20 | card reader | 4 |
| 2026-04-20 | dispenser jam | 4 |
| 2026-04-27 | cash out | 8 |
| 2026-04-27 | network | 6 |
| 2026-04-27 | card reader | 8 |
| 2026-04-27 | dispenser jam | 6 |
| 2026-05-04 | cash out | 6 |
| 2026-05-04 | network | 7 |
| 2026-05-04 | card reader | 6 |
| 2026-05-04 | dispenser jam | 4 |
| 2026-05-11 | cash out | 5 |
| 2026-05-11 | network | 8 |
| 2026-05-11 | card reader | 4 |
| 2026-05-11 | dispenser jam | 10 |
| 2026-05-18 | cash out | 7 |
| 2026-05-18 | network | 9 |
| 2026-05-18 | card reader | 7 |
| 2026-05-18 | dispenser jam | 6 |
| 2026-05-25 | cash out | 11 |
| 2026-05-25 | network | 6 |
| 2026-05-25 | card reader | 3 |
| 2026-05-25 | dispenser jam | 4 |
| 2026-06-01 | cash out | 6 |
| 2026-06-01 | network | 7 |
| 2026-06-01 | card reader | 7 |
| 2026-06-01 | dispenser jam | 2 |
| 2026-06-08 | cash out | 4 |
| 2026-06-08 | network | 6 |
| 2026-06-08 | card reader | 9 |
| 2026-06-08 | dispenser jam | 4 |
| 2026-06-15 | cash out | 5 |
| 2026-06-15 | network | 4 |
| 2026-06-15 | card reader | 6 |
| 2026-06-15 | dispenser jam | 6 |
| 2026-06-22 | cash out | 4 |
| 2026-06-22 | network | 10 |
| 2026-06-22 | card reader | 7 |
| 2026-06-22 | dispenser jam | 9 |
| 2026-06-29 | cash out | 7 |
| 2026-06-29 | network | 4 |
| 2026-06-29 | card reader | 9 |
| 2026-06-29 | dispenser jam | 4 |
| 2026-07-06 | cash out | 10 |
| 2026-07-06 | network | 7 |
| 2026-07-06 | card reader | 7 |
| 2026-07-06 | dispenser jam | 4 |
| 2026-07-13 | cash out | 8 |
| 2026-07-13 | network | 11 |
| 2026-07-13 | card reader | 6 |
| 2026-07-13 | dispenser jam | 6 |
| 2026-07-20 | cash out | 2 |
| 2026-07-20 | network | 6 |
| 2026-07-20 | card reader | 3 |
| 2026-07-20 | dispenser jam | 8 |
| 2026-07-27 | cash out | 5 |
| 2026-07-27 | network | 7 |
| 2026-07-27 | card reader | 8 |
| 2026-07-27 | dispenser jam | 5 |
| 2026-08-03 | cash out | 7 |
| 2026-08-03 | network | 5 |
| 2026-08-03 | card reader | 6 |
| 2026-08-03 | dispenser jam | 2 |
| 2026-08-10 | cash out | 8 |
| 2026-08-10 | network | 4 |
| 2026-08-10 | card reader | 7 |
| 2026-08-10 | dispenser jam | 8 |
| 2026-08-17 | cash out | 9 |
| 2026-08-17 | network | 8 |
| 2026-08-17 | card reader | 7 |
| 2026-08-17 | dispenser jam | 7 |
| 2026-08-24 | cash out | 9 |
| 2026-08-24 | network | 7 |
| 2026-08-24 | card reader | 8 |
| 2026-08-24 | dispenser jam | 7 |
| 2026-08-31 | cash out | 3 |
| 2026-08-31 | network | 7 |
| 2026-08-31 | card reader | 3 |
| 2026-08-31 | dispenser jam | 6 |
| 2026-09-07 | cash out | 7 |
| 2026-09-07 | network | 4 |
| 2026-09-07 | card reader | 2 |
| 2026-09-07 | dispenser jam | 5 |
| 2026-09-14 | cash out | 11 |
| 2026-09-14 | network | 13 |
| 2026-09-14 | card reader | 4 |
| 2026-09-14 | dispenser jam | 4 |
| 2026-09-21 | cash out | 4 |
| 2026-09-21 | network | 4 |
| 2026-09-21 | card reader | 9 |
| 2026-09-21 | dispenser jam | 5 |
| 2026-09-28 | cash out | 3 |
| 2026-09-28 | network | 3 |
| 2026-09-28 | card reader | 4 |
| 2026-09-28 | dispenser jam | 1 |


*GeoBit Large Spatial Model. Findings are labeled Evidenced (computed by a tool), Inferred (a reading or reasoning) or Unverified (no evidence yet).*