Logo Lanfrica
  • Home
  • Atlas
  • Insights
  • Docs
  • Sign in

© 2026 Lanfrica. All rights reserved. All copyrights of the resources shown on the Lanfrica website belong to the original copyright holders, unless explicitly stated otherwise.

YopaNelly/African_Market_Haggling_Simulation

Record type:

software
Creator:
Yop
Host:
A Streamlit simulation where two BDI agents a Seller and a Buyer negotiate prices in an African market. Each agent reasons logically across up to 10 rounds, with live chat bubbles showing their thinking and a real-time price chart tracking convergence toward a deal. # African Market Haggling Simulation A multi-agent negotiation simulation built with Python and Streamlit, developed as an academic assignment on Intelligent Agents. Two BDI (Belief-Desire-Intention) agents — a Seller and a Buyer negotiate the price of everyday goods in a West African market setting. Every decision is deterministic and logic-driven, and you can watch the entire negotiation unfold live, step by step, directly in the browser. --- ## What It Does The app simulates a price negotiation between two autonomous agents: - **Amara Diallo (Seller)** — knows his cost price and desired profit margin. He opens with an inflated asking price and lowers it gradually each round, but never below his minimum acceptable price. - **Fatou Ndiaye (Buyer)** — has a hard budget ceiling and a target price she considers fair. She opens with a low offer and raises it each round, but walks away if the seller stays too high for too long. The negotiation runs up to 10 rounds. Each round, both agents reason out loud — you see their Beliefs, Desires, and Intentions printed in the interface as each message appears. A live price chart shows the two prices converging (or not) over time. At the end, a banner announces whether a deal was struck or the negotiation broke down. --- ## Agent Architecture Both agents implement the **BDI (Belief-Desire-Intention)** model of rational agency: | Component | Description | |---|---| | **Belief** | What the agent knows: cost price, current offer, opponent's last move | | **Desire** | What the agent wants: profit above minimum (seller), fair price under budget (buyer) | | **Intention** | What the agent commits to doing: concede by a shrinking amount each round | Concession amounts follow a geometric decay formula: ``` concession = 0.40 × (0.60 ^ round_number) × base_gap ``` This means agents become progressively more stubborn as the negotiation advances — a realistic model of negotiation fatigue. All decisions are fully deterministic; th …

Visit

github.com

Languages

Bedawiyet