.

Author: tanmoyie

results/4.2_computational_findings_final.xlsx

@@ -24,7 +24,7 @@
     for sta in [10, 20, 30, 50]:
         try:
             instance = f'{oil} x {sta}'
-            print(f'\n\n====================================== Solving {instance} ====================================')
+            print(f'\n\n Solving {instance} ====================================')
             with open(f"../data/large scale processed dataset/preprocessed_data_o{oil}_s{sta}.pkl", "rb") as f:
                 d = pickle.load(f)
 
@@ -39,9 +39,6 @@
             M, gamma = cfg["general"]["M"], cfg["general"]["gamma"]
             nQ, nS, nH, nUN = cfg["general"]["nQ"], cfg["general"]["nS"], cfg["general"]["nH"], cfg["general"]["nUN"]
             Vehicles, c_v, C_r = cfg["assets"]["vehicles"], cfg["assets"]["c_v"], cfg["assets"]["C_r"]
-            W = [3.1, 2.5, 0.25, 0.0025, 0.025, 250, 0.25, 0.25]
-            # W = cfg["general"]["weights"]
-
             Q_vr = {(v, r): cfg["assets"]["Q_vr"][v][r] for v in cfg["assets"]["Q_vr"] for r in
                     cfg["assets"]["Q_vr"][v]}
             L_p_or = {(o, r): cfg["L_p_or_values"]["c_i"] for o in OilSpills for r in ["c", "i"]}
@@ -52,39 +49,46 @@
             spill_df = preprocess_utils.create_spill_dataframe(coordinates_spill, v_o, Sensitivity_R)
             station_df = preprocess_utils.create_station_dataframe(coordinates_st)
             DistanceMax = 10
+            # NumberStMax = 5 if sta in [10, 20] else 10
             NumberStMax = 5 if sta in [10, 20] else 10
-
+            if sta == 10: NumberStMax = 4
+            elif sta == 20: NumberStMax = 7
+            elif sta == 30: NumberStMax = 10
+            elif sta == 50: NumberStMax = 15
             #%% ------------------------------------- MILP -------------------------------------
-            print('Running MILP')
-            start_time_milp = time.time()
-            model_1, x_s, y_os, z_sor, h_sov = build_model(Stations, OilSpills, Resources, Vehicles, W,
-                                                               v_o_n, eta_o, t_os_n, gamma, M, demand_or, demand_ov, L_p_or,
-                                                               NumberStMax, A_sr, nH, nUN, nQ, Q_vr, n_vs,
-                                                               F_s, C_r, Eff_sor, pn_sor, c_v, Distance, DistanceMax, 'model_p')
-
-            model_objectives, coverage_percentage, resource_stockpile_r, x_s1, y_os1, z_sor_lamoscad, h_sov_lamoscad, solution_values, \
-                        num_var_constr = solve_model(model_1, x_s, y_os, z_sor, h_sov, OilSpills)
-            runtime_milp = round(time.time() - start_time_milp, 2)
-            print('model_objectives', model_objectives[0])
+            # print('Running MILP')
+            # W = [3.1, 2.5, 0.25, 0.0025, 0.025, 25, 0.5, 0.25]
+            # start_time_milp = time.time()
+            # model_1, x_s, y_os, z_sor, h_sov = build_model(Stations, OilSpills, Resources, Vehicles, W,
+            #                                                    v_o_n, eta_o, t_os_n, gamma, M, demand_or, demand_ov, L_p_or,
+            #                                                    NumberStMax, A_sr, nH, nUN, nQ, Q_vr, n_vs,
+            #                                                    F_s, C_r, Eff_sor, pn_sor, c_v, Distance, DistanceMax, 'model_p')
+            #
+            # model_objectives, coverage_percentage, resource_stockpile_r, x_s1, y_os1, z_sor_lamoscad, h_sov_lamoscad, solution_values, \
+            #             num_var_constr = solve_model(model_1, x_s, y_os, z_sor, h_sov, OilSpills)
+            # runtime_milp = round(time.time() - start_time_milp, 2)
+            # print('num_var_constr', num_var_constr)
+            # print('runtime_milp', runtime_milp)
             #%% ------------------------------------- Branch and Cut -------------------------------------
             # print('Running BnC')
-            # start_time_BnC = time.time()
-            # best_sol, LB_final, UB_final, obj1_from_rmp = branch_and_cut.branch_and_cut_loop(OilSpills, Stations, Resources, Vehicles,
-            #                         A_sr, C_r, Eff_sor, Distance, F_s,  v_o_n, eta_o, t_os_n, pn_sor,
-            #                         demand_or, demand_ov, nQ, Q_vr, n_vs, L_p_or, M, gamma, W, NumberStMax,
-            #                         max_iters=1, tolerance=0.01, stable_iterations=3)
-            # runtime_BnC = time.time() - start_time_BnC
-            # print("Facilities opened:", [s for s in best_sol["x"] if best_sol["x"][s] > 0.5])
-            # print("Final LB:", LB_final)
-            # print("Final UB:", UB_final)
+            W = [3.1, 2.5, 0.25, 0.0025, 0.025, 25, 0.025, 0.25]
+            start_time_BnC = time.time()
+            best_sol, LB_final, UB_final, obj1_from_rmp = branch_and_cut.branch_and_cut_loop(OilSpills, Stations, Resources, Vehicles,
+                                    A_sr, C_r, Eff_sor, Distance, F_s,  v_o_n, eta_o, t_os_n, pn_sor,
+                                    demand_or, demand_ov, nQ, Q_vr, n_vs, L_p_or, M, gamma, W, NumberStMax,
+                                    max_iters=1, tolerance=0.01, stable_iterations=3)
+            runtime_BnC = time.time() - start_time_BnC
+            print("Facilities opened:", [s for s in best_sol["x"] if best_sol["x"][s] > 0.5])
+            print("Final LB:", LB_final)
+            print("Final UB:", UB_final)
 
             results.append({
                 "instance": instance,
-                "num_var_constr": num_var_constr,
-                "model_objectives[0]": model_objectives[0],
-                "runtime_milp": runtime_milp,
-                # "milp_obj1_from_mp": round(obj1_from_rmp, 2),
-                # "runtime_BnC": round(runtime_BnC, 2)
+                # "num_var_constr": num_var_constr,
+                # "model_objectives[0]": model_objectives[0],
+                # "runtime_milp": runtime_milp,
+                "milp_obj1_from_mp": round(obj1_from_rmp, 2),
+                "runtime_BnC": round(runtime_BnC, 2)
             })
 
         except KeyError as e: