Techniques and their Applications
Dr. Amany Arafa
•Research is the process of finding the solution of
the unsolved life problems.
•Why to do research?
•Understand the world
•Solve new problems
•Improve existing solutions
• Most of the real-world problems, including design,
optimization, scheduling and control etc. are
inherently characterized by multiple conflicting
• When addressing these problems, the parameters or
variables are frequently imprecise due to
uncontrollable factors, leading to more complicated
• While solving any problem one can find multiple
• More attention must paid to :
• cost minimization.
• output maximization
• energy saving,
• environmental protection
• sustainable development issues, etc. in various fields.
• There’s lots of values to tune which are hard to predict
Optimize the solution
Set of potential solutions Optimization criteria Evaluation space
Goal: identify best solution(s) w.r.t. optimization criteria
• Nature can provide a great inspiration to the field of computation and
Nature is the only source of AI
• The beauty of nature is its ability to produce amazingly complex
patterns with amazingly simple algorithms.
• Observe how the nature work. Then Extract the principle
• Ex. Animals can find the solutions of complex problems in less energy
and cost with recovering error strategy.
• Apply natures method in solving our problem
• One of the family of insects.
• Live in tropical environment.
• Produce-cold light-chemically
• Yellow, green, pale-red light
• Based on the flashing patterns and behavior of fireflies.
Behavior of Fireflies
Two fundamental functions of such flashes are:
◦ to attract mating partners (communication)
◦ to attract potential prey
◦ protective warning mechanism
They have unique flashing pattern.
As the distance increases, light becomes weaker and
weaker because absorption by air.
Rules for Firefly Algorithm
• All fireflies will be attracted to other fireflies.
• Attractiveness is proportional to the brightness, and they both
decrease as their distance increases.
• The brightness of a firefly determined by the objective function.
Objective function f(x), x = (x1, ..., xd )
Generate initial population of fireflies xi (i = 1, 2, ..., n) Light intensity Ii at xi is
determined by f(xi)
Define light absorption coefficient
for i = 1 : n all n fireflies
for j = 1 : i all n fireflies
if(Ij > Ii), Move firefly i towards j in d-dimension; end if
Attractiveness varies with distance r via exp[−r]
Evaluate new solutions and update light intensity
end for j
end for i
Rank the fireflies and find the current best
Postprocess results and visualization.
Heuristics vs. meta-heuristics
• heuristic, is a function that ranks alternatives in search algorithms at each
branching step based on available information to decide which branch to
follow. For example, it may approximate the exact solution
• The biggest and most important difference between a heuristic and a
meta-heuristic is that heuristics get stuck in local optima, while meta-
heuristics have mechanism to avoid that.
• As examples of this mechanisms, among others, we have:
• Mutation (in Genetic Algorithm);
• Constructive with Restricted Candidate List (GRASP);
• Perturbation (ILS).
• Hyper-heuristics are to tell what sequence of Meta-heuristics to use to
solve the problem at hand. Also it can be used to classify what meta-
heuristic fits better to which problem.
• So for example, if we are solving and instance of TSP problem, it could result that the
best solution is a result of running first GA, then Simulated Annealing then Ant-
Fields Appear Based on Nature
• When using computers and computation in solving biological or
chemical problem new fields appear (bioinformatics and
• Bioinformatics is an interdisciplinary field that develops and applies
computational methods to analyze large collections of biological data.
• Cheminformatics is also known as multidisciplinary science as it
combines Chemistry, Biology, Mathematics, Biochemistry, Statistics
Application in Cheminformatics
• Storing data generated through experiments or from molecular simulation
Retrieval of chemical.
• Structures from chemical database (Software libraries).
• Prediction of physical, chemical and biological properties of chemical compounds.
• Elucidation of the structure of a compound based on spectroscopic data.
• Structure, Substructure, Similarity and diversity searching from chemical
• Docking - Interaction between two macromolecules.
• Drug Discovery
• Molecular Science, Materials Science, Food Science (nutraceuticals),
• Atmospheric chemistry, Polymer chemistry, Textile Industry,
• Combinatorial organic synthesis (COS).
Application in Bioinformatics
• The computational methods used include analytical methods,
mathematical modelling and simulation. The Nature inspired
computational methods Applied for:
• genetic sequences,
• cell populations or
• protein samples,
• to make new predictions or discover new biology.
• Gene selection
• Cancer classification
• Time tabling • Packing
• Placement • Design
• Grouping • renewable Energy design
• The optimal siting of wind turbines
• Position of solar cell
Application in Engineering
Robots Inspired by Nature
Robots as interesting complex systems
Similarity to animals
Consequences of having a real body
Real tasks in the real world --- cannot predict all interactions
Lessons learned from biological creatures
Increase physical complexity
Increase behavioral complexity
Inspiration from Insects
Exploit physical modularity
Complex robot made of simpler robots
Reconfigurable robots (Daniela Rus)
Design by Evolution
• Evolutionary Algorithms (EA) are stochastic search methods that
mimic the survival of the fittest process of natural ecosystems. The
algorithms have strong adaptability and self organization, including
Evolutionary Programming (EP) ,
• Evolutionary Strategy (ES) , Genetic Algorithm (GA),
• Differential Evolution Algorithm (DE),
• Harmony Search Algorithm (HS),
• Membrane Computing (P system),
• Teaching-learning based optimization (TLBO).
The common physics inspired algorithms include
• Chaotic Optimization Algorithm (COA),
• Intelligent Water Drops Algorithm (IWD) ,
• Magnetic Optimization Algorithm (MOA),
• Gravitational Search Algorithm (GSA) [70,71],
• Simulated Annealing (SA)
Geography inspired algorithms are
• one sort of metaheuristic algorithm and generate random solutions in
the geographical search space. These optimization algorithms are
• Tabu Search Algorithm (TS),
• Imperialistic Competition Algorithm (ICA), etc.