Research on Approximate Dynamic Programming
and
Control of Queueing Networks with Re-entrant Lines

Researchers:

José A. Ramírez-Hernández
(email: ramirejs@ececs.uc.edu)

Dr. Emmanuel Fernandez
(email: emmanuel@ececs.uc.edu)

Virtual Depository of DP and ADP Programming Tools (Password Protected)

• Matlab Code for Policy Iteration Algorithm/Application to Queueing Systems with Exp. Servers and Binary Controls.
  
  
• Matlab Code for Value Iteration Algorithm/Application to Queueing Systems with Exp. Servers and Binary Controls.
  
  
• Models in Arena with Visual Basic Code of ADP Algorithms:
  
  
  • 3-Buffer System with Q-Learning -Job Sequencing Control- (coming soon)
    
    
  • 4-Buffer System with SARSA(lambda) -Input Regultion and Job Sequencing Controls- (coming soon)
    
    
  • 4-Buffer System with Optimistic Policy Iteration and TD(lambda) and Function Approximation -Model 1, Job sequencing control-(Available)
    
    
  • 4-Buffer System with Optimistic Policy Iteration and TD(lambda) and Function Approximation -Model 2, Job sequencing control- (Available)
    
    
  • 4-Buffer System with Optimistic Policy Iteration and TD(lambda) and Function Approximation -Model 3, Job sequencing control- (Available)
    
    
  • 4-Buffer System with Optimistic Policy Iteration and TD(lambda) and Function Approximation -Model 4, Job sequencing control- (Available)

-> Click here to Return to Home Page of ADP Research