Serial Burglary Pattern Analysis

A residential area experienced 45 burglaries over 12 months with similar patterns. Students will create geographic heat maps, analyze temporal patterns, and develop predictive risk models.

# Aplly.xyz Case Study Submission

## Title
Serial Burglary Pattern Analysis

## Type
Investigative Forensics

## Difficulty
Intermediate

## Estimated Time
60 minutes

## Overview
A residential area experienced 45 burglaries over 12 months with similar patterns. Students will create geographic heat maps, analyze temporal patterns, and develop predictive risk models.

## Case Details

Function Focus: Investigative Forensics — manual reasoning, decomposition, and critical judgment (no spreadsheet or AI tool permitted in Phase 1)

Scenario:
You are the analyst at a fictional consultancy ("Praxis Advisors") tasked with answering: Serial Burglary Pattern Analysis. You have a clean, synthetic dataset described below. You must produce a defensible answer using structured reasoning — no tool-assisted shortcut on the first pass.

Dataset Structure:
- incident_id
- date
- time
- location
- area_type
- entry_method
- items_stolen
- estimated_loss_rs
- day_of_week
- distance_from_previous_km
- mo_signature
- cctv_available
- witness_present
- suspect_description
- linked_incidents

Tasks:
1. Read the dataset and reconstruct the sequence of events/evidence. State explicitly which records are confirmed facts vs. inference.
2. Identify the key anomaly or cluster that points to a primary suspect, source, or mechanism. Explain your filtering logic by hand.
3. Build a timeline or network diagram on paper linking the relevant records, showing how the evidence chain connects.
4. Weigh at least two alternative explanations (hypotheses) for the finding and use the data to eliminate one. Be explicit about what data would change your conclusion.
5. Only after completing the above manually, verify your filtering/timeline with a spreadsheet or tool and note any discrepancy.

Expected Output:
A one-page investigation memo: findings, evidence chain/timeline, eliminated hypotheses, final conclusion with confidence level.

Evaluation Criteria:
Quality of evidence-chain logic, explicit fact-vs-inference separation, correctness of hypothesis elimination, defensibility of the conclusion.

## Data Sources

| incident_id | date | time | location | area_type | entry_method | items_stolen | estimated_loss_rs | day_of_week | distance_from_previous_km | mo_signature | cctv_available | witness_present | suspect_description | linked_incidents |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| B001 | 2026-01-05 | 02:15 | Green Park | Residential | Window forced | Jewelry, cash | 850000 | Sunday | 0 | Rear window | True | False | Medium build, dark clothes | 4 |
| B002 | 2026-01-12 | 03:00 | Lajpat Nagar | Commercial | Door lock picked | Electronics | 320000 | Monday | 2.5 | Rear entry | False | False | None | 3 |
| B003 | 2026-01-19 | 02:45 | Defence Colony | Residential | Window forced | Jewelry | 1200000 | Monday | 3.8 | Rear window, jewelry only | True | True | Tall, mask | 5 |
| B004 | 2026-02-02 | 02:30 | Hauz Khas | Residential | Terrace access | Laptop, cash | 450000 | Monday | 4.2 | Upper floor | False | False | None | 3 |
| B005 | 2026-02-09 | 03:15 | Greater Kailash | Residential | Window forced | Jewelry, watches | 2200000 | Monday | 2.1 | Rear window | True | False | Medium build, gloved | 6 |
| B006 | 2026-02-16 | 01:45 | Vasant Vihar | Residential | Door forced | Cash, gold | 950000 | Monday | 5.6 | Front door, quick exit | False | True | Short, stocky | 2 |
| B007 | 2026-02-23 | 02:00 | Saket | Commercial | Window forced | Phones, cash | 180000 | Monday | 3.2 | Small window | False | False | None | 4 |
| B008 | 2026-03-02 | 02:20 | Dwarka | Residential | Balcony access | Jewelry | 680000 | Monday | 8.5 | High-rise balcony | True | False | Thin build | 5 |
| B009 | 2026-03-09 | 03:00 | Rohini | Residential | Door lock picked | Cash, electronics | 420000 | Monday | 6.8 | Rear entry | False | False | None | 3 |
| B010 | 2026-03-16 | 02:10 | Pitampura | Residential | Window forced | Jewelry | 1500000 | Monday | 3.5 | Rear window | True | True | Tall, athletic | 7 |


Full dataset: https://github.com/arora200/aplly_case_db/datasets/070-crime-serial-burglary (synthetic, 15 records)

## Solution Frameworks
Evidence chain reconstruction, hypothesis testing (deductive elimination), timeline analysis, pattern & anomaly detection, source corroboration

## Solver Guidance & Tutorials
_Solver guidance added by the pipeline (tutorial links) — see `solver_guidance` field._

## What You'll Learn
- Rebuild a chain of events from partial records
- Separate confirmed fact from inference
- Eliminate hypotheses with data

## Tags
Property Crime, Geographic Profiling, Pattern Analysis, Crime Mapping

## Registration Links
Register as Solver / Register as Evaluator