diff --git a/PSO.py b/PSO.py
new file mode 100644
index 0000000..937c35a
--- /dev/null
+++ b/PSO.py
@@ -0,0 +1,96 @@
+import random
+import Problem
+import numpy as np
+ 
+# This class contains the code of the Particles in the swarm
+class Particle:
+    def __init__(self,problem,c1,c2,w):
+        self.problem = problem
+        self.c1 = c1
+        self.c2 = c2
+        self.w = w
+        self.position = problem.generate_random_ponit()
+        self.velocity = np.random.uniform(0,1,problem.n)
+        self.personal_best = self.position.copy()
+        self.personal_best_value = 0
+
+    def update_positions(self):
+        #Update the position:
+        #position = position + velocity 
+        self.position = self.position + self.velocity   
+        return
+ 
+    def update_velocities(self, golbal_best):
+        #Update the velocity:
+        #velocity = w * velocity + c1 * r1 * (golbal_best - position) 
+        #           + c2 * r2 * (personal_best - position).
+        for i in range(self.problem.n):
+            r1 = np.random.rand()
+            r2 = np.random.rand()
+            
+            #social part
+            social = self.c1 * r1 * (golbal_best[i] - self.position[i])
+            
+            #recognition part
+            cognitive = self.c2 * r2 * (self.personal_best[i] - self.position[i])
+            
+            self.velocity[i] = (self.w * self.velocity[i]) + social + cognitive
+        return
+ 
+
+# This class contains the particle swarm optimization algorithm
+class ParticleSwarmOptimizer:
+    def __init__(self,problem,swarmSize,iterations,c1,c2,w):
+        self.problem = problem
+        self.swarmSize = swarmSize
+        self.iterations = iterations
+        self.swarm = []
+        self.generate_init_swarms(c1,c2,w)
+
+    def generate_init_swarms(self,c1,c2,w):
+        for i in range(self.swarmSize):
+            #Initialize particle
+            particle = Particle(self.problem,c1,c2,w)
+            particle.personal_best_value = self.fitness_function(particle.personal_best)
+            self.swarm.append(particle)
+        self.update_global_best()        
+        
+ 
+    #update the global best particle
+    def update_global_best(self):
+        self.global_best = self.swarm[0].position
+        self.global_best_value = self.swarm[0].personal_best_value
+        for j in range(self.swarmSize):
+        
+            #If the personal best value value is better than the global best 
+            # value Set current personal best value as the new global best
+            personal_best_value = self.swarm[j].personal_best_value
+            if personal_best_value < self.global_best_value:
+                self.global_best = self.swarm[j].personal_best
+                self.global_best_value = personal_best_value
+    
+    #Update the personal best positions
+    def update_personal_best(self):
+        for l in range(self.swarmSize):
+            
+            #If the fitness value is better than the personal best fitness
+            # value Set current value as the personal best
+            personal_best_value = self.swarm[l].personal_best_value
+            current_value = self.fitness_function(self.swarm[l].position)
+            if current_value < personal_best_value:
+                self.swarm[l].personal_best = self.swarm[l].position  
+                self.swarm[l].personal_best_value = current_value
+
+    def fitness_function(self,position):
+        return self.problem.get_value(position)
+
+    def optimize(self):
+        for i in range(self.iterations ):
+            for k in range(self.swarmSize):
+                #Update velocitie of each paricle
+                self.swarm[k].update_velocities(self.global_best)
+                #Update position of each paricle
+                self.swarm[k].update_positions()
+            self.update_personal_best()
+            self.update_global_best()
+        return self.global_best
\ No newline at end of file
diff --git a/Problem.py b/Problem.py
new file mode 100644
index 0000000..c9ec320
--- /dev/null
+++ b/Problem.py
@@ -0,0 +1,122 @@
+from abc import ABC,abstractmethod
+from enum import Enum
+import numpy as np
+import math
+#Problem interface for all test functions
+class IProblem(ABC):
+    n = 0
+    lbound = 0
+    ubound  = 0
+
+    def generate_random_ponit(self):
+        return np.random.uniform(self.lbound,self.ubound,size=(self.n))    
+
+    def generate_random_ponits(self,count):
+        return np.random.uniform(self.lbound,self.ubound,size=(count,self.n))
+   
+    @abstractmethod
+    def get_value(self, swarm):
+        pass
+
+#an enum for test functions
+class ProblemType(Enum):
+    Ackley = 1,
+    PowellSum = 2,
+    Booth = 3,
+    DropWave = 4,
+    SumSquares = 5,
+    Griewank  = 6
+
+#a factory class for generate test function class
+class ProblemFactory:
+    @staticmethod
+    def generateProblem(type: ProblemType,n,lbound,ubound):
+        if  type == ProblemType.Ackley:
+            return Ackley(n,lbound,ubound)
+        elif type == ProblemType.Booth:
+            return  Booth(n,lbound,ubound)
+        elif type == ProblemType.PowellSum:
+            return PowellSum(n,lbound,ubound)
+        elif type == ProblemType.DropWave:
+            return DropWave(n,lbound,ubound)
+        elif type == ProblemType.SumSquares:
+            return SumSquares(n,lbound,ubound)
+        elif type == ProblemType.Griewank:
+            return Griewank(n,lbound,ubound)
+class PowellSum(IProblem):
+
+    def __init__(self,n,lbound,ubound):
+        self.n = n
+        self.lbound = lbound
+        self.ubound = ubound
+    
+    #get value for custom state
+    def get_value(self, swarm):
+        value = 0
+        for i in range(self.n):
+            value = value + abs(swarm[i])**(i+2)
+        return value
+
+class Booth(IProblem):
+    def __init__(self,n,lbound,ubound):
+        self.n = n
+        self.lbound = lbound
+        self.ubound = ubound
+
+    def get_value(self,swarm):
+        value = (swarm[0] + (2 * swarm[1]) - 7)**2 + ( (2 * swarm[0]) + swarm[1] - 5)**2
+        return value
+
+class DropWave(IProblem):
+    def __init__(self,n,lbound,ubound):
+        self.n = n
+        self.lbound = lbound
+        self.ubound = ubound
+
+    def get_value(self, swarm):
+        part1 =  -1 * math.cos(12 * (math.sqrt((swarm[0]**2) + (swarm[1]**2))))
+        part2 =  (0.5 * (swarm[0]**2) + (swarm[1]**2)) + 2
+        value = part1 / part2
+        return value        
+
+class Ackley(IProblem):
+    C1=20 
+    C2=.2
+    C3=2*np.pi
+    def __init__(self,n,lbound,ubound):
+        self.n = n
+        self.lbound = lbound
+        self.ubound = ubound    
+    def get_value(self, swarm):
+        part1 = -1. * self.C1 * np.exp(
+            -1. * self.C2 * np.sqrt((1./self.n) * sum(map(lambda nb: nb**2, swarm)))
+            )
+        part2 = -1. * np.exp(
+            (1./self.n) * \
+            sum(map(lambda nb: np.cos(self.C3 * nb), swarm))
+            )
+        return part1 + part2 + self.C1 + np.exp(1)
+
+class SumSquares(IProblem):
+    def __init__(self,n,lbound,ubound):
+        self.n = n
+        self.lbound = lbound
+        self.ubound = ubound
+    def get_value(self, swarm):
+        return np.sum(np.arange(1,self.n+1) * (swarm**2))
+
+class Griewank(IProblem):
+    def __init__(self,n,lbound,ubound):
+        self.n = n
+        self.lbound = lbound
+        self.ubound = ubound    
+    def get_value(self, swarm):
+        part1 = 0
+        for i in range(len(swarm)):
+            part1 += swarm[i]**2
+        part2 = 1
+        for i in range(len(swarm)):
+            part2 *= math.cos(float(swarm[i]) / math.sqrt(i+1))
+        return 1 + (float(part1)/4000.0) - float(part2)   
+
+
diff --git a/README.md b/README.md
index c411466..ac1df71 100644
--- a/README.md
+++ b/README.md
@@ -1 +1 @@
-# PSO
\ No newline at end of file
+#Particle Swarm Optimization (PSO) with Python
\ No newline at end of file
diff --git a/main.py b/main.py
new file mode 100644
index 0000000..5124963
--- /dev/null
+++ b/main.py
@@ -0,0 +1,17 @@
+import PSO
+import Problem
+import numpy as np
+
+ 
+problem = Problem.ProblemFactory.generateProblem(
+                Problem.ProblemType.Ackley, 5, -32, 32)
+                
+iterations = 100
+swarmSize = 50
+c1 = 2 #social constant
+c2 = 2 #cognative constant
+w =  0.2 #constant inertia weight
+pso = PSO.ParticleSwarmOptimizer(problem, swarmSize, iterations, c1, c2, w)
+solution = pso.optimize()
+print('global best position : {0} - global best value : {1}'.format(solution,problem.get_value(solution)))
+            
\ No newline at end of file