I am looking for new methods and new trends in optimization, which have absorbed many interests. I will be grateful to you if you can give me some information.
Over the last years, one of the trends is the algorithms based on column generation. I'm working mainly on vehicle scheduling and there are more and more serious publication with this technique.
There are diversified categories of heuristic and meta-heuristic methodologies which have been dynamically studied. Of meta-heuristic type methods, Harmony search and Firefly algorithm might be considered of somehow more recent approaches, however, there are various population based approaches which have shown promising ability in optimization problems such as Ant Colony, Bee Colony, Particle Swarm Optimization (PSO as one of most prominent ones) while Simulated annealing and Travelling salesman problem might be constructive. Applied Soft Computing as a leading journal in this field can further clarify this subject with detailed information.
Actually, this is a very large-scope question. From my experience, I have faced many problems which required different optimization algorithms. To name a few, Dynamic Programming, Pontryagin's Minimum Principle, Genetic Algorithm, Particle Swarm Optimization. For stochastic systems, Stochastic Dynamic Programming plays a signification role. For complex systems with large state space and large control space, Approximate Dynamic Programming (ADP) is your choice. I think, ADP is very interesting and there has been a significant research on it in the past few years.
One of the recent success stories is hybridizing metaheuristics and mathematical programming. Details can be found under the acronym MATHEURISTICS. (See, e.g., http://en.wikipedia.org/wiki/Matheuristics). We have a conference on this (in Hamburg, Germany) which just opened the submission page:
http://iwi.econ.uni-hamburg.de/mh14/
Moreover there are special tracks at various conferences on this topic; see, e.g. the GOR Meeting in Aachen, Germany: http://www.or2014.de/
This question does not have a simple answer. Optimization is a field that is far too big for that. Depending on the type of optimization problem you face there are many new emerging technologies. Just to provide a rough classification:
You may want to find local optima or global optima
the optimization problems may be unconstrained or with various types of constraints
There are finite dimensional optimization problems (low dimensional or high dimensional) or infinite dimensional problems
for the finite dimensional ones you can have
combinatorial problems (integer problems)
continuous problems
mixed-integer problems
for every one of these you can have
special structure (linear, quadratic, etc.)
factorable problems (i.e. all functions are known by explicit formulas)
problems for which semi-local or global information is available (Lipschitz constants, convexity, automatic estimates)
problems for which only local information is available (point evaluations, derivative evaluations, etc.)
there it makes a difference, how many derivatives you get (Hessians y/n, gradients y/n?)
it also makes a huge difference, how much effort one function evaluation takes (expensive or cheap functions)
For infinite dimensional optimization problems there are also many classes
semi-infinite problems
optimal control problems
problems with differential equation constraints
For all of these problem classes there are various new emerging techniques available, depending on whether you want to find your solution with certainty (complete algorithms), provably with probability 1 eventually (asymptotically complete algorithms), or very likely (incomplete algorithms). If you provide more information on your problems, we might provide more specific information.
Which is the latest Optimization technique for Power system problem solving? - ResearchGate. Available from: https://www.researchgate.net/post/Which_is_the_latest_Optimization_technique_for_Power_system_problem_solving2 [accessed Nov 16, 2015].