100% found this document useful (1 vote)

1K views

Routing and Scheduling Algorithms

The document discusses routing and scheduling problems, which involve planning the most efficient routes for vehicles to service locations while meeting time window constraints. It describes several types of routing problems like the traveling salesman problem (TSP), vehicle routing problem (VRP), and pickup and delivery problem with time windows (PDPTW). It also outlines approaches for constructing initial routes using methods like nearest neighbor and Clark-Wright savings, as well as improving routes through techniques like 2-opt and 3-opt exchanges. The goal is to minimize costs like distance, time or number of vehicles while satisfying capacity limits and time windows for customer locations.

Uploaded by

Timothae Zacharyy

Available Formats

Download as PDF, TXT or read online on Scribd

100% found this document useful (1 vote)

1K views

Routing and Scheduling Algorithms

Uploaded by

Timothae Zacharyy

Available Formats

Download as PDF, TXT or read online on Scribd

You are on page 1/ 25

E211 - Operations Planning II

SCHOOL OF ENGINEERING E211 – OPERATIONS PLANNING II

Routing and Scheduling Problems
The scheduling of customer service and the routing of
service vehicles are at the heart of many service
operations.
Service routing: dispatching of installation or repair technicians

Passenger routing: routing and scheduling of school buses,

public buses, ambulances

Freight routing: pickup and delivery service of logistics firms;

postal and parcel delivery

Scheduling involves planning the timing for each

location to be visited.

Routing involves forming the sequence in which

locations are to be visited

SCHOOL OF ENGINEERING E211 – OPERATIONS PLANNING II

Objectives and constraints of Routing and
Scheduling problems

Minimize total cost of service delivery or route by one or

more of the following:
Minimize total distance traveled
Minimize total time traveled
Minimize number of vehicles utilized
Minimize number of personnel on duty

Subject to the following considerations:

Vehicles
Capacities
Release, maximum and down times

Customers
Time windows
Priority
Pickup and delivery

SCHOOL OF ENGINEERING E211 – OPERATIONS PLANNING II

Characteristics of Routing and Scheduling Problems

Often can be represented using network diagrams

Consist of nodes and arcs
Nodes represent locations to be visited. Depot node is the
begin/end location (For today’s problem, ETI is the depot
node).
Arcs represent the possible paths among the locations. Arcs
can be directed (one-way, represented by arrow) or undirected
(two ways, represented by line)

Route – sequence in which nodes are to be visited

Schedule – time for each node to be visited

SCHOOL OF ENGINEERING E211 – OPERATIONS PLANNING II

Feasibility of a Tour

A route is also known as a tour. E.g. 1 2 3 4 5 1

For a tour to be feasible,
1. It must start and end with the depot node.
2. All nodes must be included.
3. A node must be visited only once.

SCHOOL OF ENGINEERING E211 – OPERATIONS PLANNING II

Classification of Routing and Scheduling Problems

Most commonly seen routing and scheduling problems

can be classified into the following categories:

Traveling salesman problem (TSP)

Multiple traveling salesman problem (MTSP)

Vehicle routing problem (VRP)

Vehicle routing problem with time windows (VRPTW)

Pickup and delivery problem with time windows (PDPTW)

SCHOOL OF ENGINEERING E211 – OPERATIONS PLANNING II

Travelling Salesman Problem

Given a set of towns and the distances between them, the TSP
determines the shortest tour starting from a given town,
passing through all the other towns exactly once and returning
to the first town.
Type of decisions
Routing

SCHOOL OF ENGINEERING E211 – OPERATIONS PLANNING II

Multiple Travelling Salesman Problem

The MTSP is a generalization of the TSP, where more than

one salesman (vehicle) is needed to form the tours due to the
limit on time (distance) travelled by each salesman.
In MTSP, all the salesman (vehicle) are to leave from and
return to a common depot.
Typical applications of MTSP are in service routing, where, for
example, dispatching of more than one installation or repair
technician is needed.
Type of decisions
Routing
Assigning

SCHOOL OF ENGINEERING E211 – OPERATIONS PLANNING II

Vehicle Routing Problem
In VRP, a fleet of vehicles located at a central depot has to serve
a set of geographically dispersed customers and return to depot.

Each vehicle has a given capacity and each customer has a given
demand.

The objective is to minimize the total cost (travelling distance) of

serving the customers.

Typical applications of VRP is the delivery of goods located at a

central depot to customers who have placed orders for such goods.
Type of decisions
Assigning
Routing

SCHOOL OF ENGINEERING E211 – OPERATIONS PLANNING II

Vehicle Routing Problem with Time Windows

The VRPTW is similar to the VRP except that, each

customer may require to be served within a given time
window.

Type of decisions

Assigning
Routing
Scheduling
Tour 1

Tour 2

SCHOOL OF ENGINEERING E211 – OPERATIONS PLANNING II

Pickup and Delivery Problem with Time Windows

The PDPTW is similar to the VRP except that, in addition

to the size of the load to be transported, each customer
service also specifies
The location where the load is to be picked up and the pickup time
window;
The location where the load is to be delivered and the delivery
time window;

Type of decisions

Assigning Depot

Routing
Scheduling

The PDPTW determines how to form the routes and schedule the pickup and delivery.

SCHOOL OF ENGINEERING E211 – OPERATIONS PLANNING II

Solution approaches to the Routing and Scheduling
Problems
Exact algorithms such as integer programming algorithms
For a problem with n drop-off points, there are n! possible routings.
E.g. 8 drop-off points 40320 possible routings!
The solution space increases extremely fast as the number of drop-
off points increases, and therefore exact algorithms are often not
suitable for large size problems.

Heuristic algorithms
Rule-of-thumb methods
May not achieve optimality but solutions are often reasonably good.

Metaheuristics
Top-level heuristics guiding other heuristics to search for better
feasible solutions in the solution space, such as Genetic Algorithms,
Ant Colony Optimization, Simulated Annealing, Tabu search.

They are very commonly used in solving complex vehicle routing and
scheduling problems.

SCHOOL OF ENGINEERING E211 – OPERATIONS PLANNING II

Solution approaches to the Routing and Scheduling
Problems
Commonly used vehicle routing heuristics
Construction algorithm: an algorithm that determines a tour
according to some construction rules, but does not try to
improve upon this tour. Examples are Nearest Neighbor method
and Clark and Wright Savings method.

Improvement algorithm: an algorithm that performs a

sequence of edge or vertex exchanges within or between vehicle
routes to improve the current tour. Examples are 2-opt, 3-opt,
1-relocate, 2-swap.

Two-phase algorithm: an algorithm that constructs vehicle

routes in two phases
Cluster-first-route-second method: customers are first organized
into feasible clusters, and a vehicle route is constructed for each of
the clusters.
Route -first-cluster-second method: a tour is first built on all
customers and the tour is then segmented into feasible vehicle
routes.

SCHOOL OF ENGINEERING E211 – OPERATIONS PLANNING II

Tour construction—Nearest Neighbor Method

Select a node 1 (depot) to start with.

Add the closest node to node 1.

Repeat by adding to the last node the closest unvisited

node until no more nodes are available.

Connect the last node with the first node.

SCHOOL OF ENGINEERING E211 – OPERATIONS PLANNING II

Tour construction—Clark and Wright Savings Method

Select a node 1 (depot).

Compute the savings, Sij for linking nodes i and j.
Sij = c1i + c1 j − cij for i and j = nodes 2, 3, ..., n.

where cij = the cost of travelling from node i to node j.

Rank the savings from the largest to the smallest.
Starting from the top of the list, form larger subtours by linking
‘appropriate’ nodes i and j (so that j is visited immediately after i
on the resulting route). Stop when a complete tour is formed.

Here ‘appropriate’ means:

Not violating route constraints such as vehicle capacity or time window constraints; and
i and j fulfil one of the following conditions:

SCHOOL OF ENGINEERING E211 – OPERATIONS PLANNING II

Tour construction—Clark and Wright Savings Method

i and j fulfil one of the

following conditions:

A. Neither i nor j has already

been assigned to a route.
Initiate a new route including
both i and j
Adjacent to Depot

B. Exactly one of the two

points (i or j) has already been
included in an existing route
and that point is adjacent to
the depot. Add link (i, j) to that
same route.

C. Both i and j have already

been included in two different
existing routes and both of
them are adjacent to the
depot. Merge the two routes.

SCHOOL OF ENGINEERING E211 – OPERATIONS PLANNING II

Tour improvement methods: 2–opt

2–opt
A 2-opt move consists of eliminating two edges and reconnecting
the two resulting paths in a different way to obtain a new tour.

In implementation, 2-opt moves are performed iteratively until no

more improvements can be found through 2-opt moves.

Reconnect Remove

Original 2–opt After 2–opt

Remove Reconnect

SCHOOL OF ENGINEERING E211 – OPERATIONS PLANNING II

Tour improvement methods: 3–opt

3–opt
A 3-opt move consists of eliminating three edges in a tour and
reconnecting the tour in all other possible ways, and then
evaluating each reconnection method to find the optimum one.

In implementation, 3-opt moves are performed iteratively until

no more improvements cane be found through 3-opt moves.

SCHOOL OF ENGINEERING E211 – OPERATIONS PLANNING II

Tour improvement methods: 1–relocate

1–relocate
A 1-relocate move consists of eliminating one node in a tour and
reinsert it into the tour in all other possible ways, and then
evaluating each relocation method to find the optimum one.