liveBook · Manning

Pheromone evaporation

As explained in Optimization Algorithms: AI techniques for design, planning, and control problems

Pheromone evaporation is the reduction of simulated pheromone concentration over time to weaken older search information.

Pheromone evaporation reduces simulated pheromone concentration over time, weakening older search information in ant colony optimization (ACO). It is one phase of pheromone updating; pheromone deposit is the other. section 10.2 section 10.6

The evaporation rate controls the balance between following established paths and exploring alternatives:

  • With a rate of 0, trails persist indefinitely, encouraging repeated selection of one path and risking entrapment in a local optimum. section 10.2
  • With a rate of 1, trails disappear at the maximum rate, forcing random exploration and potentially slowing convergence. section 10.2
  • Intermediate settings let better paths attract ants while allowing older information to fade and other paths to remain discoverable. section 10.2 section 10.6

Ant foraging

Several ants begin at a nest with equal pheromone values on two paths, so each path initially has equal selection probability. After traversal, evaporation weakens existing trails before new pheromone is deposited. Repeating this process can strengthen promising paths without permanently fixing the colony to its earliest choice. section 10.2

Figure 10.4 Effect of evaporation rate during the food foraging process. In the simulation, the ants initiate their search for food from the central nest, which is surrounded by three food sources shown as blobs. The pheromone trails are shown in white. Upon discovering a food item, an ant transports it back to the nest, leaving behind a chemical trail. This trail is then followed by other ants that pick up the scent, directing them toward the food source. As more ants continue to retrieve food, they strengthen the chemical trail.

Use the foraging simulation to observe how different evaporation rates change trail persistence, path following, and exploration. section 10.2

How Optimization Algorithms: AI techniques for design, planning, and control problems covers it

How deep it goes
Defines the evaporation phase, names its rate, and contrasts extreme settings without deriving the update equation.
Examples
Uses equal-pheromone two-path ants and a NetLogo food-foraging simulation to illustrate trail persistence.
What you can do after
Operate/configure evaporation in an ACO or NetLogo foraging simulation and anticipate its exploration–convergence trade-off.
How it connects
Connects pheromone deposit with exploration, exploitation, local optima, and ant colony optimization.

Also explained in

Compare how 2 books explain pheromone evaporation →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Optimization Algorithms: AI techniques for design, planning, and control problems. The section links point to the parts of the book it draws on.