An Integrated Model for Production and Distribution Planning of Perishable Products with Inventory and Routing Considerations

Algorithm 2

Details of particle swarm heuristic.

Run LINGO to find an initial feasible solution

Set the initial feasible solution as Localbest Solution and Globalbest Solution

Generate particles and set all equal to initial feasible solution

Set all the particles equal to initial feasible solution

While the stopping criterion is not met

  For each particle 𝑋, calculate 𝑉=𝑉+𝛼⋅(𝑋_{𝑙𝑜𝑐𝑎𝑙𝑏𝑒𝑠𝑡}−𝑋)+𝛽⋅(𝑋_{𝑔𝑙𝑜𝑏𝑎𝑙𝑏𝑒𝑠𝑡}−𝑋).

  For each bit 𝑥^𝑗 in particle 𝑋, If (rand < Sigmoid(𝑣^𝑗)) then 𝑥^𝑗 = 1, else 𝑥^𝑗 = 0

  Check feasibility for each particle and repair the particle

  Calculate fitness function for each particle

  Update Localbest Solution and Globalbest Solution

End while

Run LINGO using Globalbest Solution to obtain real valued variables