Telecommunications and optimization

The telecommunications industry is young and dynamic. All over the world, the deregulation of telecommunications markets lead to a wide range of new companies, which are competing among each other. The last decade has been marked by fast growth and rapid introduction of new technologies and services.

In this competitive market, it is vital to rationalize the workflow and to automate the processes. Optimization becomes a key technology.

Today, optimization is often seen as a post-planning process, responsible for improving a given situation. To maximize the overall profit, however, optimization must take a more global perspective. It must be fully integrated into all strategic and operational decisions, and in particular, optimization technologies should already be embedded within the initial planning stages.

Over the last decade, computational power and optimization technologies evolved. Numerous cooperations between industry and scientific research have shown that state-of-the-art optimization technology can significantly reduce costs, increase capacity or service quality in telecommunication networks.

Benefits of optimization

There are manifold benefits of optimization for telecommunication service providers.

Assessment of solution quality

Modern optimization technologies add a new quality to the planning process. In addition to the computation of network configurations, they provide information to assess the quality of a solution. It becomes possible to give serious answers to the question how bad a decision could be in the worst-case, or on the positive side, to see that one took an excellent or even an optimal decision.

High-quality planning results

With automated intelligent optimization algorithms it is possible to search, evaluate and compare numerous solutions in short computation times. The consequence are high-quality planning results. The planning process itself and its results in particular become reproducable and more reliable. Aside from this, the dependency of planning departments from employee turnover is reduced.

Reduced planning time

Today, many planning tasks are solved in a semi-automatic or manual process. Since optimization is accompanied by automation, all its advantages are added in such a situation. In particular, part of the planning can be performed in the background, the overall planning time decreases and, in consequence, the total cost of planning can substantially be reduced.

Methods of optimization

There is a wide range of technologies to optimize a process or a planning task. Depending on development and planning requirements, atesio selects together with its customers the appropriate solution approach among the following alternatives:

Exact optimization

Very efficient technologies exist among the exact optimization algorithms which allow to compute provably optimal solutions. The most successful is mixed-integer programming. It is based on strong theoretical improvements from the last fifty years and it is particularly well-suited for many types of decision and configuration problems. Based on a thorough investigation of the planning task, these methods typically end in optimization algorithms which allow to compute excellent network configurations within short computation times. As unique feature, the quality of solutions becomes assessable.

Fast and efficient heuristics

Many planning tasks are performed on a regular basis and are based on extensive evaluation of planning alternatives. In such a situation, atesio provides fast and efficient heuristics. Typically, these are a combination of construction heuristics and local search procedures which allow to compute good quality solutions in the range of seconds or minutes.

Randomized local search algorithms

In some planning scenarios, extensive computational effort is permitted but the time to develop customized optimization algorithms is strongly limited. In this case, atesio suggests to apply randomized local search algorithms such as its popular representatives tabu search, simulated annealing, and genetic algorithms. These allow to compute good quality solutions given a time frame of a night or a week-end.

Optimization with atesio

As spin-off of ZIB, one of the world-wide leading institutes in mathematical optimization, atesio is putting state-of-the-art optimization technology into action. The many years of experience to develop applications for industrial partners make atesio the perfect partner to get the maximum out of your network resources.

© 2002-2016 by atesio GmbH Impressum