Ants may be common insects, but they have some unique talents. There are more than 10,000 species of ants, which belong to the order Hymenoptera, which includes wasps and bees. Ants are social creatures and typically live in structured nest communities in ground-level mounds, underground or in trees.
Carpenter ants nest in wood and can be destructive to buildings; some species, such as army ants, do not have permanent homes and hunt for food for their huge colonies during periods of migration.
Ant communities are headed by a queen or queens, whose role is to lay thousands of eggs to ensure the colony’s survival. Worker ants are females, wingless and never reproduce, but forage for food, look after the queen’s offspring, work on the nest and protect the community.
Male ants often have the one role, to mate with the queen, after which they usually die. Ants communicate and cooperate using chemicals that can alert others to danger or a source of food. Ants normally eat nectar, seeds or insects, but army ants can prey on reptiles, birds and small mammals.
One ant cannot do much on its own, but the colony as a whole solves complex problems, like building a sophisticated nest, maintaining it and filling it with food.
Computer scientists and games creators have been on the search for artificial intelligence (AI) for decades and AI based on the behaviour of ants is having a surprising degree of success.
Their behaviour interested people like Marco Dorigo, a researcher at the Free University of Brussels and one of the founders of the field that has become known as swarm intelligence.
Dr Dorigo was interested that ants are good at choosing the shortest possible route between a food source and their nest, something reminiscent of the classic computational conundrum, the traveling salesman problem.
Given a list of cities and their distance apart, the salesman must find the shortest route to visit each city once, which gets more complicated as the number of cities increases. A computer trying to solve it will take longer and longer and suck in more and more processing power.
The traveling salesman problem is interesting because many other complex problems, such as designing silicon chips and assembling DNA sequences, come down to a modified version of it.
The ants’ solution is to use chemical signals called pheremones. When an ant finds food, she takes it back to the nest, leaving behind a pheremone trail that will attract others. The more that follow it, the stronger the trail will become. The pheremones evaporate quickly and once all the food had been collected, the trail goes cold.
The speed of evaporation makes long trails less attractive than short ones and pheremones amplify the limited intelligence of single ants into something more powerful.
In 1992, Dr Dorigo and his team started developing Ant Colony Optimisation (ACO), an algorithm that looks for solutions to a problem by simulating a group of ants wandering over an area and laying down pheremones. ACO was good at solving traveling-salesman-type problems and has grown into a family of algorithms applied to many practical questions, most successfully in logistics.
The Swiss supermarket chain, Migros, and Italian pasta-maker, Barilla, both manage their daily deliveries from central warehouses to local retailers using AntRoute, a programme developed from ACO principles. Each morning the software calculates the best routes and delivery sequences according to all the variables of the day. It takes 15 minutes to produce a delivery plan for 1,200 lorries.
Ant-like algorithms have also been applied to routing information through communication networks. Dr Dorigo and his colleague, Gianni Di Caro, have developed AntNet, a routing protocol in which packets of information hop from node to node, leaving a trace that signals the ‘quality’ of their trip as they do so and other packets sniff the trails created and choose accordingly.
The costs of implementing such routing protocols can be prohibitive for large corporations, but ant routing looks particularly promising for ad hoc mobile networks, such as those used by the armed forces and civil protection agencies.
The global telecommunications company, BT, has been researching ants for years to try and keep busy telephone networks from jamming, using the ants’ techniques of self-organisation, which does away with a single over-arching control for whole systems.
Fabrice Saffre, principal researcher at BT’s impressively named Pervasive ICT Research Centre, says: “With self-organisation you have very simple rules governing individual units that together perform a bigger task. The simplicity makes for less computation and is easier on the network.”
Saffre says: “It’s a very economical solution, especially for very dynamic problems. Anything you can do with self-organisation is basically a free lunch.”
BT’s Embryo is a biologically inspired programme that allows self-organising behaviour among servers. As tasks are allocated among servers, each analyses its load and if it is required by too many users for a task, it will send a message to its nearest neighbours in the network to discover which has that application available.
If those neighbours do not have the application, they contact their nearest neighbours and so on. When a server with the application is found, the task is reallocated. Each server independently balances its load and only communicates with the servers to which it is directly connected, rather than the whole network.
The overall effect on the network is ‘load balancing’, a fair and effective system that is more efficient than a rigid control system, as well as being more robust to failures. If you were to have a rigid control system that depended on a fixed and even distribution of tasks network-wide, it could fall apart if any part of the network is lost.
Dr Dorigo is also working on something that can act as well as think; robots. A swarm of small, cheap robots can achieve, through co-operation, the same results as individual big and expensive robots and with more flexibility and robustness; if one robot falls, the swarm keeps going.
Eye-bots could look around and locate interesting objects, foot-bots give hand-bots a ride to the location and the hand-bot picks up the object of interest and they all run home. He believes bot-swarms could be used for surveillance and rescue, perhaps locating and saving survivors during a fire.
Ndubuisi Ekekwe is the founder of the African Institution of Technology and wanted to build a networked digital library storing theses from African universities. He was having problems until he studied a group of swarming ants at work.
They were constantly re-organising themselves and transporting bits of food 30 times their size. He took the decision to learn from their behaviour. They worked as a team, they trusted each other and were open partners, diligent and focused.
He decided to do away with the notion that only by working alone could he ensure quality and formed a team for his project, bringing professionals together, giving each assignments based on capability. The team would keep working, even slowly, with deadlines for focus and Ekekwe would be open to new ideas if something was not working.
The project was a success and Ekekwe says: “Ants can teach us a great deal about planning, military strategy and business management. They make decisions together as a group and depend on one another to survive.”
There are some key lessons for business owners in this. By engaging and trusting everyone in the organization, you can achieve greater success. Leaders must not think only they can close sales, install products and fine-tune designs. They should give others the opportunity to succeed or fail and always ask for help.
A team leader should share a project’s progress with the team; when people know where they are they can come up with suitable solutions, you never know who may have the information or networks to unlock future growth opportunities. Ekekwe says: “There is an ancient saying: ‘The ant-hills are not built by elephants, but by the collective efforts of the little rejected ants.’”
Humans make comparisons between choices, which can lead to irrational decisions, whereas individual ants typically know of only one option, which prevents them from making potentially misleading comparisons. Their ‘wisdom of the crowds’ is essentially teamwork, which can be the key to preventing the kind of mistakes individuals are prone to make when working alone.