TİKDAS — Climate Resilience Platform — Case Study

Problem

Turkey had no digital system to measure agricultural enterprises' resilience to climate change — assessments were manual, non-reusable and could not be aggregated across household, regional and national levels.

Solution

Built a single Spring Boot backend powering both a researcher web tool and a Flutter mobile self-assessment app, implementing a 13-indicator resilience algorithm across social, economic, environmental and managerial dimensions. Validated with 922+ field surveys across 8 strategic crops and 8 Turkish provinces.

Architecture

$ render architecture.mmd

flowchart TB
  subgraph Field[Field]
    Farmer[Farmer · Flutter App]
  end
  subgraph Lab[Research Lab]
    Researcher[Researcher Web Tool]
  end
  Farmer --> API[Spring Boot API]
  Researcher --> API
  API --> Algo[13-Indicator<br/>Resilience Algorithm]
  Algo --> Score[Resilience Score]
  API --> Survey[Survey Validator]
  Score --> DB[(PostgreSQL)]
  Survey --> DB
  DB --> Export[CSV · Excel · JSON Export]

Technical decisions

$ git log --oneline decisions/

#01

One backend, two clients

The same Spring Boot API and resilience algorithm served both the researcher web tool and the farmer mobile app — no two implementations to drift.

#02

13-indicator algorithm as a pure function

The scoring logic lived in a deterministic, side-effect-free service so academic reviewers could replay any historical assessment from raw inputs.

#03

Multi-format export from one query model

CSV, Excel and JSON exports projected from the same read model so researchers always got matching numbers across formats.

Have a similar challenge?

Let's talk