TS.m

function [costs, bestSol] = TS(jobs, m, n, listLength, iterations, costFunc, getBestNeighbor)
  % This function implements the tabu search algorithm for Hybird Job Shop Sceduling

  costs = [];
  tabuList = zeros(listLength, 2);

  % Firstly, randomly generate an initial solution given m and n
  % Set the best solution to the initial solution
  bestSol = ones(1, n);
  for i = 1 : m
    bestSol(i) = randi([1,m],1);
  end
  bestSolCost = costFunc(bestSol, jobs, m, n);
  costs(1)=bestSolCost;

  sol = bestSol;
  for i = 2 : iterations
    % Get the best solution in the neighbourhood of the current solution
    % avoiding Tabu moves
    [sol, solCost, tabuList] = getBestNeighbor(sol, jobs, m, n, tabuList, ...
                                               bestSolCost, costFunc);
    %disp(tabuList);
    costs(i)=solCost;

    %Update the best solution
    if (solCost<bestSolCost)
      bestSol = sol;
      bestSolCost = solCost;
    end
  end
end