Portuguese 
English 
Spanish 
4 min of reading 
0 comments 
The Technique Uses Ant Colony Optimization to Test Thousands of Combinations and Cut Material While Maintaining Safety Standards in Structural Design
The crossing of Argentine ants along two paths between the nest and food became the basis for a powerful computing idea. From the chemical trail left on the ground, a method was born that today helps to seek out more cost-effective and efficient concrete bridge designs.
This path led to Ant Colony Optimization (ACO) algorithms. From the 1990s onwards, the technique began to be used in routing, logistics, and also in fine-tuning real structures, such as bridge piles.
Two Bridges, One Nest, and a Discovery That Became Technology
In classic tests, a nest was connected to a food source by two bridges. In some situations, both paths were of equal length; in others, one was shorter.
-
The architect who dreamed of this futuristic bridge died ten years before seeing it completed, and now, with a single 200-meter mast holding almost a kilometer of structure, it has just opened as the largest in the world in its category.
-
Gas concessionaires, sanitation companies, and other services will no longer be able to tear up the concrete highways of Santa Catarina to lay pipes, and those who cause damage will have to rebuild the road with the same original material and thickness, with no expiration date.
-
Just over 20 meters of concrete are missing for Brazil and Paraguay to be physically connected for the first time over the Paraguay River on a nearly 1.3 km bridge that will open an unprecedented road shortcut between the Atlantic and the Pacific and shorten the journey of goods to Asia.
-
Hong Kong built 1,800 housing units in just 12 months using prefabricated modules in China, while the same type of traditional public housing would take an average of 4.1 years. The United Court project was the first fully completed transitional housing in the city.
At first, the ants spread out and chose paths almost at random. As they walk, they leave a trail of pheromones, a chemical mark that can be followed by other ants.
Over time, the faster path gets more traffic and accumulates more pheromone. This increases the likelihood that new ants will choose the same route, reinforcing a dominant path without any central command.
Pheromone on the Ground Becomes a “Shortcut” Without Anyone Directing
The mechanism is simple and efficient. The more ants pass a point, the stronger the signal becomes, and the more appealing it is to the rest of the group.
This reinforcement creates a repetition effect. The shortest path tends to be taken more often in less time, which accelerates the accumulation of pheromone and consolidates the choice.
The result is a sort of natural highway that emerges from individual actions but delivers a robust collective solution.
ACO Creates Digital Pheromone and Makes the Computer “Learn” the Best Path
The idea was transformed into an algorithm in the early 1990s. Instead of real insects, artificial ants come into play, exploring possible alternatives for a problem.
Each virtual ant constructs a solution by choosing steps, routes, or design combinations. When a solution improves, the algorithm deposits more digital pheromone on the decisions that led to that outcome.
At the same time, part of this virtual pheromone gradually disappears, like evaporation. This prevents the system from getting stuck in old answers and keeps space to test new possibilities until it converges to near-ideal options.
The Algorithm Enters the Calculation of Hollow Bridge Pillars
In structural design, a concrete bridge involves many variables. Considerations include the geometry and dimensions of the piles, concrete choices, amount of steel, and foundation details, all adhering to safety requirements.
Testing all combinations through trial and error is not feasible. The solution is to let the algorithm explore the design space and seek configurations that reduce costs without deviating from performance and strength regulations.
In a study using this approach, 15 hollow rectangular piles were optimized. Heights ranged from 20 to 60 meters for viaducts with spans of 40 to 60 meters, with the algorithm tailored to the structural problem.
Savings Over 20% in Concrete and Steel Without Losing Safety
The application of ACO to this type of pile allowed for material reduction. The cut in concrete and steel was over 20% compared to traditional methods, while maintaining strength and service requirements.
Less material also means less associated production impact. The gain appears in reduced CO₂ emissions and a decrease in total cost, especially when the viaduct consists of multiple piles.
To reach this point, the process was executed with populations of around 50 virtual ants for approximately 100 stages, totaling hundreds or thousands of iterations until stabilizing on optimal solutions within constraints.
The Future of Concrete Could Be Determined by “Swarms” of Algorithms
ACO is part of a group of bio-inspired techniques used to solve challenging optimization problems. Its good performance in routing, logistics, and structures shows how simple rules can turn into powerful engineering tools.
These methods also combine with other strategies, such as genetic algorithms, threshold searches, and simulated annealing, to automate stages and accelerate design choices.
The same logic that guides the traffic of ants on the ground can influence decisions that reduce costs and materials in large concrete structures while meeting the project’s technical requirements.
The practical application is clear: observing natural behavior helped create a tool capable of seeking more efficient designs, with the potential for savings when the method is applied to viaducts with many piles.
-
-
-
-
-
10 people reacted to this.