Live: tarimsalkoridor.online · Region: Demre · Finike · Kaş · Kemer · Kumluca · Envelope: 36.18°N–37.12°N / 29.30°E–30.85°E
Agri-DSS turns local agronomic and economic knowledge into a decision a farmer can act on. Pick a district and a neighborhood, and the system returns a concrete, defensible plan: the seasonal crops most likely to deliver yield and profit, a long-term orchard/tree investment suited to the land, and an emerging market opportunity to watch — all printable on a single A4 sheet for a village board or cooperative.
It is built for the Western Antalya agricultural corridor (Demre, Finike, Kaş, Kemer, Kumluca): five districts, 147 neighborhoods, each with a curated recommendation profile.
Farmers acting in isolation tend to plant whatever looked profitable last season — everyone plants tomatoes at once. The predictable result at harvest is oversupply, a price collapse below cost, wasted produce, and a losing year for the very growers who worked hardest.
Agri-DSS is designed to go beyond "what's the weather?" and "when do I fertilize?". Its goal is to recommend, for each neighborhood, the crop that optimizes profitability — weighing not only soil and climate but also expected market prices and (in a future phase) what other growers in the region are already planting. The question it answers:
"If I plant this, will I get both high yield and a strong harvest-time price — and therefore maximum profit?"
- Dynamic region selection — filter by District → Neighborhood across a clean, guided flow.
- Seasonal recommendations — the best soil-based crops for the selected area, each rated for estimated yield, estimated profitability, and an agronomic/economic rationale.
- Long-term investment advice — a tree/orchard crop matched to the land and climate (avocado, olive, pomegranate, almond, geographically-protected Finike orange, and more).
- Conceptual market analysis — an emerging demand / supply-gap opportunity for the area.
- Printable report — a typographic A4 layout ready for village boards and cooperatives.
Agri-DSS is a fully client-side static site — no backend, no build step. This makes it trivially hostable (GitHub Pages), instantly auditable, and impossible to break with a server outage.
index.html → Interface + all DSS logic (custom Swiss-style UI, vanilla JS)
data.json → Data layer — content is decoupled from code
CNAME → Custom domain (tarimsalkoridor.online)
LICENSE → Apache-2.0
The interface follows the Swiss / International Typographic Style: a strict modular grid, strong typographic hierarchy (Archivo + Space Mono), generous negative space, and a single decisive accent — agricultural green on warm paper. The experience is interactive (guided stepper, corridor diagram, staggered result reveals, live counters) yet collapses to a clean, ink-on-white A4 document when printed.
Data is stored in a compact, DRY structure so the catalogue can be edited without touching code:
| Key | Purpose |
|---|---|
cropSets |
Reusable seasonal crop sets (e.g. seraDomatesBiberSalatalik) |
longTermCrops |
Dictionary of long-term investment crops (e.g. avokado, zeytin) |
regions |
District → Neighborhood → record |
Each neighborhood record carries either a cropSet reference or an inline crops list;
either a longTerm reference or an inline longTermCustom; plus a marketGap string.
index.html fetches data.json on load and resolves these references at runtime.
To update recommendations, edit
data.jsononly — no code changes required.
Because data.json is loaded via fetch, open the site through a web server rather than
double-clicking the file:
# from the repo root:
python3 -m http.server 8000
# then open http://localhost:8000- Push to the
mainbranch. - Settings → Pages → Source: Deploy from a branch →
main/root. - The root
CNAMEbindstarimsalkoridor.online; point your DNS records at GitHub Pages.
There is no build pipeline — index.html and data.json are published as-is.
The current dataset is conceptual, derived from regional reports (economy, environment,
land use, climate). It demonstrates the system's logic and interface. In production, the same
data.json contract can be backed by real soil, climate, and market data sources — without any
change to the application code.
Released under the Apache-2.0 license. See LICENSE for details.