diff --git a/kattis/subway2.py b/kattis/subway2.py
new file mode 100644
index 0000000..1656114
--- /dev/null
+++ b/kattis/subway2.py
@@ -0,0 +1,76 @@
+# https://open.kattis.com/problems/subway2
+
+from math import dist
+from collections import defaultdict
+from math import inf  # Default value in Dijkstra's algorithm
+from sys import stdin  # Instead of input
+
+adjacency_list = defaultdict(dict)
+line_dict = defaultdict(dict)
+
+TRAIN_SPEED = 40_000 / 60  # meter / min
+WALK_SPEED = 10_000 / 60  # meter / min
+
+# Parse home and school coordinates
+data = list(stdin)
+first_line = list(map(int, data[0].split()))
+home = tuple(first_line[:2])
+school = tuple(first_line[2:])
+
+# Parse subway station's coordinates, add them to the adjacency_list and calculate their weights
+for row in data[1:]:
+
+    # Parse
+    subway = list(map(int, row.split()))[:-1]
+    subway = [*zip(subway[::2], subway[1::2])]
+
+    # Add surrounding stations with the travel time to the adjacency_list
+    sis = set(range(len(subway)))
+    for station_index, station in enumerate(subway):
+
+        # Add the station before and after (if possible)
+        for i in [-1, 1]:
+
+            if station_index + i in sis:
+                line_dict[station][subway[i + station_index]] = dist(station, subway[i + station_index]) / TRAIN_SPEED
+
+
+# Add all the walkingways between stations. But only if there doesn't already exist a connection.
+all_stations = list(line_dict) + [home, school]  # Add the start- and endpoint to the graph
+for current_station_index, current_station in enumerate(all_stations):
+    adjacency_list[current_station] = {s: dist(s, current_station) / WALK_SPEED
+                               for s in all_stations[:current_station_index] +
+                                       all_stations[current_station_index + 1:]}  # Dict comprehension
+    adjacency_list[current_station].update(line_dict[current_station])  # Replace walkways with subways
+
+# Initialise some variables for the algorithm
+visited = set()
+node_information = {x: {"dist": inf, "from": None} for x in adjacency_list}
+node_information[home]["dist"] = 0
+
+# Using Dijkstra's algorithm to calculate the shortest path from home to school
+current = home  # Start at home
+while True:
+
+    if current not in visited:
+        visited.add(current)
+        stack = []
+
+        for neighbor in adjacency_list[current]:
+
+            if neighbor not in visited:
+                new_dist = node_information[current]["dist"] + adjacency_list[current][neighbor]
+
+                if new_dist < node_information[neighbor]["dist"]:
+                    node_information[neighbor] = {"dist": new_dist, "from": current}
+                stack.append((node_information[neighbor]["dist"], neighbor))
+
+        if stack:
+            current = min(stack)[1]
+
+        else:
+            break
+
+
+# Print result! Woohoo!
+print(int(round(node_information[school]["dist"])))