Genetic algorithms are implemented as a computer simulation in which a population of abstract representations (called chromosomes or the genotype or the genome) of candidate solutions (called individuals, creatures, or phenotypes) to an optimization problem evolves toward better solutions more


Genetic Algorithms


Definition: A genetic algorithm is a search technique that uses the principles of evolution and natural selection theory in order to optimise business processes and best apportion limited resources.

The biological theory of natural selection has long been an accepted truth for biologists. The theory explains how species evolve through natural selection – selective breeding based on desired and undesired genetic characteristics.

In biology, the success or failure of any single organism to produce offspring depends largely on how the genetic characteristics of the organism relate to the environment in which it lives. While a characteristic may be advantageous under one set of circumstances, the same characteristic may become detrimental if the environment changes.

Genetic Algorithms in Business

The same theory applies to business processes. While a process may be ideally designed to deal with current market conditions and the needs of the enterprise, if those needs or conditions change the process will no longer be optimal.

Genetic algorithms can be used to optimise business processes by considering the demands and conditions of the surrounding environment - essentially, allowing process designers to design a process that will be most likely to succeed in the current market environment.

Methodology of Genetic Algorithms

Process designers can optimise their business processes by following a simple process to identify optimal conditions.

* Generate a Range of Potential Solutions (Individuals)

genetic algorithmInitially, it is necessary for process designers to identify a number of different methods by which a process could be completed. Following the metaphor of a biological system, each of these solutions is known as an ‘individual’ (or, according to MIT, a genome).

The initial ‘population’ could be made up of anything from just a few individuals to hundreds or even thousands, depending on the scope of the process and the variety of ways in which it can be performed. Traditionally, these individuals will be generated randomly from throughout the entire search space of the process. The search space is a theoretical construct in which is contained every possible individual, no matter how unlikely.

* Evaluation

Each individual must then be evaluated to produce a fitness function. Fitness functions refer to a quantifiable scalar-valued measure of the fitness of an individual. In the case of business processes the fitness functions of each process will most likely be generated by evaluating the time it would take to complete each process. Faster processes would receive high fitness functions, and slower ones would receive lower functions.

* Selection

Once each individual has been evaluated, the individuals with the highest fitness functions will be combined to produce a second generation. In general the second generation of individuals can be expected to be ‘fitter’ than the first, as it was derived only from individuals carrying high fitness functions.

* Termination

This process of selection and reproduction should, ideally, allow process designers to develop an ever-fitter solution, exploiting the collected advantages of its predecessors.

The process of recombining and reproducing individuals will continue until a predefined condition has been met, such as:

A solution has met the minimum predefined criteria.

Selection and reproduction no longer results in increased fitness with each subsequent generation (suggesting that a fitness plateau has been reached).

Financial and technical budget has been reached.

Benefits of Genetic Algorithms

Benefits of genetic algorythmsOne of the problems with the manual design of business processes is the fact that the process can only be as good as the imagination of the designer. Genetic algorithms, however, allow designers to evaluate every possible method of performing a process that exists within the search space. While many of these solutions will be impractical (since search space contains every solution that is theoretically possible, regardless of practicality), evaluating the search space will ensure that all possible methods are considered.

Perhaps the greatest value of genetic algorithms is in the fact that they are based on the theory of ever-evolving optimisation as a response to changing environments. There can be few more apt metaphors for a successful, adaptable enterprise than this. Genetic algorithms therefore offer enterprises the tools necessary to achieve supreme agility and adaptability.

In order to stay ahead of the competition it may be advisable to use this method periodically to test how processes measure up to current market conditions. If the fitness function of any process falls below a predefined rate the process can be re-evaluated. As a performance management tool, genetic algorithms are second to none.