Everything you need to run the A.D.I. platform is already in the repository. The only tools required are Git (to clone the repo) and Power BI Desktop (free from Microsoft). The dashboard fileDocumentation Index
Fetch the complete documentation index at: https://mintlify.com/HelenDiMo/archaios-data-Intelligence/llms.txt
Use this file to discover all available pages before exploring further.
wwii_analisys_dashboard.pbix sits at the repository root and is pre-configured to load the three ADI master CSV files automatically from the same directory — there is no database to configure, no cloud service to connect, and no environment to provision. Clone the repository, open the .pbix file, and the full interactive dashboard is ready to use.
Prerequisites
Git
Any recent version of Git is sufficient. Used to clone the repository to your local machine. Available for Windows, macOS, and Linux.
Power BI Desktop
Free desktop application from Microsoft. Required to open
wwii_analisys_dashboard.pbix and explore the interactive campaign dashboards. Download from the official Microsoft page.Installation
Clone the repository
Open a terminal and run the following commands to clone the repository and navigate into the project directory:This downloads the complete project, including the Power BI dashboard file, all three ADI master datasets, the EDA notebook, and the five raw THOR source files in
/data/.Verify the data files are present
Confirm that the three ADI master CSV files are in the repository root before opening Power BI:These three files are the data sources that the Power BI report reads directly at load time:
If any of these files are missing, the Power BI report will fail to load its data. Re-clone the repository to restore the complete file set.
| File | Contents |
|---|---|
adi_dataset_master.csv | Consolidated master dataset spanning the full 1943–1944 campaign window |
adi_master_1943_1944_CLEAN.csv | Cleaned and enriched phase dataset for the 1943–1944 strategic panorama view |
adi_master_1_5_junio.csv | Phase dataset for the June 1–5, 1944 pre-D-Day logistical softening period |
Open the dashboard
Launch Power BI Desktop, then use File → Open report to open:The file is located at the repository root. Power BI will automatically resolve the three CSV data source connections from the local directory. No manual data source reconfiguration is required provided the CSV files remain in the same directory as the
.pbix file.Once loaded, use the interactive map to explore bombing density across the Normandy theater, and use the slicer panels to filter by date range, aircraft series, and target city.Explore the three campaign views
The report contains three pages, each corresponding to a distinct phase of the Allied air campaign. Navigate between them using the page tabs at the bottom of the Power BI window:1943–1944 — Strategic Panorama
The full-campaign overview across the European Theater of Operations. Identifies macro-level target concentrations, mission frequency trends, and high-bomb-density zones across the entire 1943–1944 period. The starting point for understanding the strategic context before narrowing to the Normandy theater.1 A 5 JUNIO — Pre-D-Day Logistical Softening
A detailed breakdown of the five days immediately preceding the D-Day landings (June 1–5, 1944). This view is analytically critical for archaeological prospection: it shows the intense targeting of inland transport infrastructure — bridges, railway marshalling yards, and road junctions — that Allied air planners used to isolate the Normandy beaches from German reinforcement. Bomb tonnage in these inland zones was massive and is a primary indicator of buried structural remains.DIA-D — June 6, 1944
A visual record of D-Day itself: the geographic distribution of Allied air missions on June 6, 1944. Documents the coastal assault support missions, beach-area targets, and the broader air umbrella that covered the amphibious landings — one of the most operationally significant single days of aerial warfare in history.
Working with the Jupyter Notebook
Data analysts who want to inspect, re-run, or extend the EDA pipeline can work directly with thedesembarco_normandia.ipynb notebook. The notebook requires a standard Python data science stack — no specialist dependencies beyond the libraries listed below.
Install the dependencies and launch the notebook with:
/data/ directory, performing data loading, dimension inspection, type checking, null analysis, and iterative cleaning passes for each file. It is the canonical record of every data transformation decision made in the pipeline — from raw THOR extract to the finished ADI master CSV files that power the dashboard.
The notebook outputs include printed schema summaries (row counts, column types, null counts) and sample rows for each source dataset, making it straightforward to audit data quality at every stage of the pipeline even without re-executing the cells.
Repository Structure
The complete repository layout after cloning is as follows:/data/ directory holds the five raw and pre-cleaned THOR source files — the unmodified primary source layer. The three adi_*.csv files in the repository root are the enriched ADI master datasets produced by the notebook and consumed by Power BI. The .pbix file and the master CSVs must always share the same directory.
Next Steps
Understand the Data Model
Explore the structure of the three ADI master tables, their column definitions, data types, and the relationships that drive the Power BI semantic model.
Navigate the Dashboard
A guided walkthrough of all three campaign-phase dashboard views — filters, map interactions, KPI cards, and how to read the bomb-density visualisations for archaeological targeting.
Run the EDA Notebook
Step-by-step documentation of the
desembarco_normandia.ipynb pipeline: data loading, cleaning decisions, enrichment logic, and how to extend the analysis with new THOR data.Data Quality & Biases
A critical assessment of THOR’s four known data biases — representativeness, geographic coverage, target labelling, and temporal recording quality — and their implications for archaeological decision-making.