NetLogo User Community Models

(back to the NetLogo User Community Models)

Download If clicking does not initiate a download, try right clicking or control clicking and choosing "Save" or "Download".(The run link is disabled because this model uses extensions.)

## WHAT IS IT?

This is an Agent-Based Model (ABM) that simulates a competitive public tender process. It is designed to explore how complex bidding behaviors and market dynamics, such as a "race to the bottom," emerge from the interactions of individual firms (agents) with different strategies and capabilities.

## HOW IT WORKS

The simulation consists of a series of tender rounds where player agents compete for contracts.

**Player Agents:**
Each agent represents a firm with a unique set of attributes:
- **Archetype:** Players are assigned an archetype ("aggressive," "conservative," "adaptive," or "follower") that governs their bidding behavior and risk tolerance.
- **Experience & Quality:** Players have an experience level and a base quality capability, which influence their performance and how they are evaluated.
- **Profit Targeting:** Each player has an internal target profit margin they aim to achieve.

**Bidding Process:**
1. **Ideal Bid:** A player first calculates an "ideal bid" based on their estimated cost for the tender and their target profit margin.
2. **Strategic Adjustments:** This ideal bid is then heavily adjusted based on a `bid-strategy` value, which is learned over time. Players adapt this strategy by observing the market and the success of others.
3. **Final Bid:** The final bid is a result of these adjustments, reflecting a balance between the player's ideal profit and the competitive reality of the market.

**Evaluation & Learning:**
- **MEAT Criteria:** Tenders are awarded using the Most Economically Advantageous Tender (MEAT) criteria, which is a weighted combination of **price**, **quality**, and **experience**. You can adjust the weights of these criteria using the sliders.
- **Social Learning:** Players learn from their peers. They observe the performance of others—not just who wins, but who wins *profitably*—and may imitate the strategies of the most successful players. This drives the evolution of strategies in the market.

## HOW TO USE IT

**Buttons:**
- **setup:** Resets the simulation and creates the players and evaluators.
- **go:** Runs the simulation one round at a time.
- **simulate-n-rounds:** Runs the simulation for the number of rounds specified by the `number-rounds` slider.

**Key Sliders:**
- **number-players:** Sets the number of competing firms.
- **number-rounds:** Sets the duration of the simulation.
- **meat-price-weight, meat-quality-weight, meat-experience-weight:** Adjust the importance of each factor in the tender evaluation. The model will automatically normalize these so they sum to 1.
- **social-learning-rate / strategy-imitation-threshold:** Control how quickly and easily players imitate each other's strategies.

**Key Plots:**
- **Bid Trends (Ideal vs Actual):** This plot is key to observing emergent behavior. It shows the average "ideal bid" (what players *want* to bid for profit) versus the average "actual bid" (what they bid after competitive adjustments). The divergence between these lines reveals the pressure of the market.
- **Win Rates by Archetype:** Shows the relative market share (percentage of total wins) held by each player archetype over time.
- **Profit Margins by Archetype:** Tracks the average profit margin achieved by each archetype.

## THINGS TO TRY

- **The Race to the Bottom:** Run a simulation for 100 rounds. Watch the "Bid Trends (Ideal vs Actual)" plot. Does a gap form between the ideal and actual bids? This shows the competitive pressure forcing players to abandon their profit targets.
- **Varying MEAT Criteria:** Set the `meat-price-weight` to be very high (e.g., 0.8) and run the simulation. Which archetype tends to dominate? Now try again with `meat-quality-weight` set high.
- **Market Competition:** Run a simulation with a low `number-players` (e.g., 5) and another with a high number (e.g., 40). How does the level of competition affect the "race to the bottom"?
- **Social Dynamics:** Turn the `strategy-imitation-threshold` down to make imitation very easy. Does one strategy quickly take over the entire market?

## EXTENDING THE MODEL

- **Market Shocks:** Introduce sudden changes to the simulation, such as a sudden increase in the `BASE-TENDER-VALUE` or a change in the MEAT weights mid-run.
- **Player-Specific Knowledge:** Give players imperfect information. For example, some players might have better `cost-estimation-accuracy` than others.
- **Coalitions:** Allow players to form temporary alliances to bid on larger, more complex tenders.

## CREDITS AND REFERENCES
Made by Riccardo Pizzuti
Mail: r.pizzuti92@gmail.com
Linkedin: https://www.linkedin.com/in/ricpiz92/
GitHub: https://github.com/RicPiz