diff --git a/16/part2.py b/16/part2.py
new file mode 100644
index 0000000..40b390d
--- /dev/null
+++ b/16/part2.py
@@ -0,0 +1,82 @@
+import re
+
+def getInput(filename):
+	with open(filename, "r") as f:
+		fields, myTicket, nearbyTickets = f.read().split("\n\n")
+		myTicket = myTicket.replace("your ticket:\n", "")
+		nearbyTickets = nearbyTickets.replace("nearby tickets:\n", "")
+
+		return (
+			{m[1]: m[2] for m in re.finditer(r"(.+): ([^\n]+)", fields)},
+			[int(n) for n in myTicket.split(",")],
+			[[int(n) for n in ticket.split(",") ] for ticket in nearbyTickets.split("\n")]
+		)
+
+class FieldRule:
+	def __init__(self, raw: str):
+		self.raw = raw
+		match = re.match(r"(\d+)\-(\d+) or (\d+)\-(\d+)", raw)
+		self.range1 = [int(match[1]), int(match[2])]
+		self.range2 = [int(match[3]), int(match[4])]
+
+	def __call__(self, num: int):
+		return (self.range1[0] <= num <= self.range1[1]) or \
+					 (self.range2[0] <= num <= self.range2[1])
+
+	def __repr__(self):
+		return f'FieldRule("{self.raw}")'
+
+def isFieldValid(tickets, fieldRule, pos):
+	for ticket in tickets:
+		if not fieldRule(ticket[pos]):
+			return False
+	return True
+
+# A backtracking algorithm to find the field order
+def solveFieldOrder(tickets, fields: dict):
+	fieldOrder = {}
+
+	numOfFields = len(fields)
+	unassignedFields = fields.copy()
+	while len(fieldOrder) < numOfFields:
+		possibleFields = {}
+
+		for fieldName, fieldRule in unassignedFields.items():
+			for i in range(numOfFields):
+				if i not in fieldOrder.values() and isFieldValid(tickets, fieldRule, i):
+					possibleFields[fieldName] = possibleFields.get(fieldName, [])
+					possibleFields[fieldName].append(i)
+
+		for fieldName, orders in possibleFields.items():
+			if len(orders) == 1:
+				fieldOrder[fieldName] = orders[0]
+				del unassignedFields[fieldName];
+
+	return fieldOrder
+
+fields, myTicket, nearbyTickets = getInput("input.txt")
+
+# Preproccess the ticket so that checking is faster
+for fieldName, fieldRule in fields.items():
+	fields[fieldName] = FieldRule(fieldRule)
+
+# Filter out the invalid tickets
+validNearbyTickets = []
+for ticket in nearbyTickets:
+	isTicketValid = True
+	for num in ticket:
+		if all(not rule(num) for rule in fields.values()):
+			isTicketValid = False
+			break
+	if isTicketValid:
+		validNearbyTickets.append(ticket)
+
+# Find combination of fields that is valid
+fieldOrder = solveFieldOrder(validNearbyTickets, fields)
+
+solution = 1
+for fieldName, order in fieldOrder.items():
+	if fieldName.startswith("departure"):
+		solution *= myTicket[order]
+
+print(solution)
\ No newline at end of file
diff --git a/16/test.txt b/16/test1.txt
similarity index 100%
rename from 16/test.txt
rename to 16/test1.txt
diff --git a/16/test2.txt b/16/test2.txt
new file mode 100644
index 0000000..c629bd3
--- /dev/null
+++ b/16/test2.txt
@@ -0,0 +1,11 @@
+class: 0-1 or 4-19
+row: 0-5 or 8-19
+seat: 0-13 or 16-19
+
+your ticket:
+11,12,13
+
+nearby tickets:
+3,9,18
+15,1,5
+5,14,9
\ No newline at end of file