Bruce L. Golden R.H. Smith School of Business University of Maryland Presented at DSI Annual Meeting November 2013 Baltimore, MD 1.

Slides:



Advertisements
Similar presentations
Impact of Interference on Multi-hop Wireless Network Performance Kamal Jain, Jitu Padhye, Venkat Padmanabhan and Lili Qiu Microsoft Research Redmond.
Advertisements

1 Material to Cover  relationship between different types of models  incorrect to round real to integer variables  logical relationship: site selection.
An Exact Algorithm for the Vehicle Routing Problem with Backhauls
DOMinant workshop, Molde, September 20-22, 2009
The Variable Neighborhood Search Heuristic for the Containers Drayage Problem with Time Windows D. Popović, M. Vidović, M. Nikolić DEPARTMENT OF LOGISTICS.
Solving the dispatching problem by MILP MIMIC (Mixed Integer Maurizio Igor Carlo) Carlo Mannino SINTEF ICT (Oslo), University of Rome Maurizio Boccia University.
EMIS 8373: Integer Programming Valid Inequalities updated 4April 2011.
Vehicle Routing & Scheduling: Part 1
Vehicle Routing & Scheduling Multiple Routes Construction Heuristics –Sweep –Nearest Neighbor, Nearest Insertion, Savings –Cluster Methods Improvement.
1.224J Recitation #4 Freight transportation. Topics Homework questions Home Depot MVRP: Multi vehicle routing problem – Applications – Formulation – Heuristics.
Planning under Uncertainty
The Min-Max Split Delivery Multi- Depot Vehicle Routing Problem with Minimum Delivery Amounts X. Wang, B. Golden, and E. Wasil INFORMS San Francisco November.
MAE 552 – Heuristic Optimization Lecture 24 March 20, 2002 Topic: Tabu Search.
Localized Techniques for Power Minimization and Information Gathering in Sensor Networks EE249 Final Presentation David Tong Nguyen Abhijit Davare Mentor:
Vehicle Routing & Scheduling
Erasmus Center for Optimization in Public Transport Bus Scheduling & Delays Dennis Huisman Joint work with: Richard Freling and.
Vehicle Routing & Scheduling: Part 2 Multiple Routes Construction Heuristics –Sweep –Nearest Neighbor, Nearest Insertion, Savings –Cluster Methods Improvement.
Job Shop Reformulation of Vehicle Routing Evgeny Selensky University of Glasgow
1.3 Modeling with exponentially many constr.  Some strong formulations (or even formulation itself) may involve exponentially many constraints (cutting.
1 Chapter-4: Network Flow Modeling & Optimization Deep Medhi and Karthik Ramasamy August © D. Medhi & K. Ramasamy, 2007.
A Parallel Cooperating Metaheuristics Solver for Large Scale VRP’s Jiayong Jin, Molde University College, Norway
1 1 1-to-Many Distribution Vehicle Routing John H. Vande Vate Spring, 2005.
Compact Routes for the Min-Max K Windy Rural Postman Problem by Oliver Lum 1, Carmine Cerrone 2, Bruce Golden 3, Edward Wasil 4 1. Department of Applied.
1 A Guided Tour of Several New and Interesting Routing Problems by Bruce Golden, University of Maryland Edward Wasil, American University Presented at.
Optimization Problems - Optimization: In the real world, there are many problems (e.g. Traveling Salesman Problem, Playing Chess ) that have numerous possible.
Heuristic Optimization Methods Tabu Search: Advanced Topics.
Notes 5IE 3121 Knapsack Model Intuitive idea: what is the most valuable collection of items that can be fit into a backpack?
Logical Topology Design
The Application of The Improved Hybrid Ant Colony Algorithm in Vehicle Routing Optimization Problem International Conference on Future Computer and Communication,
A Hybrid Genetic Algorithm for the Periodic Vehicle Routing Problem with Time Windows Michel Toulouse 1,2 Teodor Gabriel Crainic 2 Phuong Nguyen 2 1 Oklahoma.
Bekkjarvik, A Heuristic Solution Method for a Stochastic Vehicle Routing Problem Lars Magnus Hvattum.
K. Panchamgam, Y. Xiong, B. Golden*, and E. Wasil Presented at INFORMS Annual Meeting Charlotte, NC, November 2011 *R.H. Smith School of Business University.
Thursday, May 9 Heuristic Search: methods for solving difficult optimization problems Handouts: Lecture Notes See the introduction to the paper.
Mobile Agent Migration Problem Yingyue Xu. Energy efficiency requirement of sensor networks Mobile agent computing paradigm Data fusion, distributed processing.
The Colorful Traveling Salesman Problem Yupei Xiong, Goldman, Sachs & Co. Bruce Golden, University of Maryland Edward Wasil, American University Presented.
A Computational Study of Three Demon Algorithm Variants for Solving the TSP Bala Chandran, University of Maryland Bruce Golden, University of Maryland.
The Generalized Traveling Salesman Problem: A New Genetic Algorithm Approach by John Silberholz, University of Maryland Bruce Golden, University of Maryland.
Optimizing Pheromone Modification for Dynamic Ant Algorithms Ryan Ward TJHSST Computer Systems Lab 2006/2007 Testing To test the relative effectiveness.
Transportation Logistics Professor Goodchild Spring 2011.
Integer Programming (정수계획법)
Log Truck Scheduling Problem
Biologically Inspired Computation Ant Colony Optimisation.
Vehicle Routing Problems
Management Science 461 Lecture 3 – Covering Models September 23, 2008.
IE 312 Review 1. The Process 2 Problem Model Conclusions Problem Formulation Analysis.
1 Vehicle Routing for the American Red Cross Based on a paper by Dr. Jinxin Yi Carnegie Mellon University Published in MSOM Winter 2003,
The Minimum Label Spanning Tree Problem: Illustrating the Utility of Genetic Algorithms Yupei Xiong, Univ. of Maryland Bruce Golden, Univ. of Maryland.
Balanced Billing Cycles and Vehicle Routing of Meter Readers by Chris Groër, Bruce Golden, Edward Wasil University of Maryland, College Park American University,
Metaheuristics for the New Millennium Bruce L. Golden RH Smith School of Business University of Maryland by Presented at the University of Iowa, March.
The Computerized Routing of Meter Readers over Street Networks: New Technologies Turn Old Problems into New Opportunities by Robert Shuttleworth, University.
1 Chapter 5 Branch-and-bound Framework and Its Applications.
Joint Routing and Scheduling Optimization in Wireless Mesh Networks with Directional Antennas A. Capone, I. Filippini, F. Martignon IEEE international.
Project Coordination Model Alexander Galenko (presenter), AgileAssets, Inc Eric Perrone, AgileAssets, Inc Tonya Scheinberg, AgileAssets, Inc The 6th International.
1 Last Time Buys and Reuse at Océ A last time buy decision tool for reusable service parts. Anke Verbaarschot 2011, October 19 th.
Online Routing Optimization at a Very Large Scale
Routing Through Networks - 1
Si Chen, University of Maryland Bruce Golden, University of Maryland
1 The Travelling Salesman Problem (TSP) H.P. Williams Professor of Operational Research London School of Economics.
1.3 Modeling with exponentially many constr.
Heuristics Definition – a heuristic is an inexact algorithm that is based on intuitive and plausible arguments which are “likely” to lead to reasonable.
Integer Programming (정수계획법)
TransCAD Vehicle Routing 2018/11/29.
The Hierarchical Traveling Salesman Problem
1.3 Modeling with exponentially many constr.
Planning the transportation of elderly to a daycare center
Integer Programming (정수계획법)
Generating and Solving Very Large-Scale Vehicle Routing Problems
Tools to Address Challenges in Local Authority Waste Management
A Neural Network for Car-Passenger matching in Ride Hailing Services.
Presentation transcript:

Bruce L. Golden R.H. Smith School of Business University of Maryland Presented at DSI Annual Meeting November 2013 Baltimore, MD 1

Outline of Talk  Some opening remarks  The CETSP over a street network  Arc routing with meanderable streets  Vehicle routing with customer preference for visit order  Additional topics of interest  Conclusions 2

Opening Remarks  I have worked on vehicle routing problems since 1974  As a researcher  As an owner of RouteSmart from 1980 to 1998  Remarkable advances since 1974  This represents a major success story for OR 3

The CETSP over a Street Network  Until recently, utility meter readers had to visit each customer location and read the meter at that site  Now, radio frequency identification (RFID) technology allows the meter reader to get close to each customer and remotely read the meter  In previous work (Shuttleworth et al., 2008), our models were based on data from a utility and used an actual road network with a central depot and a fixed radius r for the hand-held device  Our goal was to minimize distance traveled or elapsed time 4

The CETSP over a Street Network  We used RouteSmart (RS) with ArcGIS  Real-world data and constraints  Address matching  Side-of-street level routing  Solved as an arc routing problem  Our heuristic selected segments to exploit the “close enough” feature of RFID  RS routed the meter reader over the chosen segments to obtain a cycle  RS solved the problem as a CPP or a RPP 5

Heuristic Implementation  How did we chose the street segments to feed into RS?  We tested several heuristic ideas  Greedy Approaches  IP Formulations  The focus was on exploiting the power of RFID in order to find a shorter route 6

Shuttleworth et al. Results  We presented several heuristics for solving this new class of problems  The best heuristics seemed to work well  RFID travel paths had a 15% time savings and 20% distance savings over the RS solution  As the technology improves (i.e., the radius increases) the savings will continue to increase 7

An Example from RouteSmart  Shortly after our work on this topic, RS developed its own commercial capability  An illustration is provided on the next few slides  So far, the focus has been on improving one route at a time, but partitioning a region into routes is also important 8

9 A Neighborhood on a Route

A Traditional Route through a Neighborhood 10

An RFID Route through the same Neighborhood 11

RFID Impact on Route Miles % - 51% - 21% - 46%

RFID Impact on Route Time 13

Designing Partitions 14 Partitions with r=100Partitions with r=150

Arc Routing with the Meander Option  Suppose there is demand for service at homes on a street  If the street is narrow and the traffic is light, it is possible (and often desirable) to service both sides of the street in a single pass (i.e., meander in one direction)  If the street is wide and traffic is heavy, we must service each side on a different pass (i.e., meandering is not allowed) 15

Meanderable Streets  In intermediate cases, we can ask the algorithm to decide which option is best – these streets are called meanderable  This is an important real-world issue  Home delivery of newspapers  Trash collection  Local delivery (e.g., UPS & Fedex)  Meter reading (for now)  Maybe USPS delivery  Irnich (2005, 2008, 2008) has studied this problem and transformed it to an asymmetric traveling salesman problem, but, otherwise, it has attracted little attention 16

Mixed Windy RPP with the Meander Option  Consider a street connecting a and b  Streets on which there is no demand are not required  For streets where there is demand on only one side, a single pass over a directed arc is required 17

Mixed Windy RPP with the Meander Option  For streets where there is demand on both sides, there are three possibilities  If Meander = No, we have two directed arcs between a and b  If Meander = Yes, we have one undirected edge between a and b  If Meander = Maybe, we have one of the above two scenarios 18

Solving the Problem Using IP  Zhang & Ming (2013) formulated this problem as an IP  It differs from Irnich’s IP, but takes about the same amount of time to solve small instances  Real-world instances were provided by RouteSmart  Zhang & Ming solved an instance with 684 nodes, 4938 arcs, 20 components, and 240 meanderable streets in 145 seconds using CPLEX

Sensitivity Studies  In general, we expect  M ij + T ij > Max {S ij + T ij, S ji + T ji } and  M ij + T ij < (S ij + T ij ) + ( S ji + T ji )  Zhang & Ming studied the impact of the number and costs of meanderable streets  As the number of meanderable streets increases, total cost tends to decrease  As the meander cost to service cost ratio increases, we meander less 20

The Importance of Meandering  We observe that even when the ratio R is large, it still might make sense to meander M ij + T ij (S ij + T ij ) + (S ji + T ji )  On a real-world instance that we solved, we found one meander with R = and another with R =  So, the meander cost can be relatively high and yet still offer cost saving opportunities 21 R=

Future Work  There is much work to be done on both exact and heuristic approaches  A commercial sanitation client asked whether we can design algorithms that take time of day into account  It may be desirable to meander some streets in the early morning (4 to 5 am), but not later 22

Vehicle Routing with Customer Preference for Visit Order  Service companies visit customer’s homes for inspections, installations, repairs, etc.  E.g., cable TV companies  A customer is informed that he will be visited on Tuesday, between 9 am and 5 pm  For some customers, that is fine  Other customers might be willing to pay an extra amount to be visited early or late in the day 23

Customer Preference for Visit Order  Given that it may be impossible to estimate the duration of a service call with precision, it makes more sense to ask customers to pay extra to be visited first, second, last, next to last, etc. on a route  Two approaches  Set a price in advance (e.g., $25, $15, and $5) for first, second, and third on a route  Allow customers to bid (or not) for visit order  The goal is to minimize {travel cost – revenue} 24

Initial Progress on this Problem  Sahin, Golden, Raghavan (2013) have begun to study this problem  We start with a TSP version  One service technician can visit n customers per day  We considered two MILP formulations 25

MILP Formulation  A modified Dantzig (1963) formulation  It has on the order of n 3 binary variables 1 if the technician travels from i to j  x ijt = and visits j in order t 0 otherwise  It is rarely used to solve the TSP  A modified Miller-Tucker-Zemlin (1960) formulation  It has on the order of n 2 binary variables 26

Numerical Study  10 instances of 20 customers (n=20) each  Coordinates generated randomly in a 100 x 100 square  Distances are Euclidean  20%, 30%, or 40% of the customers place bids  They bid for the first and last 3 or 5 positions  Bids are generated using a Normal distribution  The two formulations are solved using CPLEX

 The MTZ formulation seems sensitive to the percentage of customers bidding  The Dantzig formulation shows no such sensitivity 28 Formulation Comparison Bids20%30%40% FormulationDantzigMTZDantzigMTZDantzigMTZ B & B Nodes , ,105, ,871,894 CPU Time (s) Average LP-IP Gap13.02%46.51%12.62%58.48%9.91%63.44%

Future Work 29  We have managed to solve (to optimality) instances with 50 customers for the TSP version and 80 customers for a VRP version (both with bidding)  The VRP version assumes there are K vehicles and that each vehicle services exactly Q customers (i.e., KQ = n)  There is much work to be done on both exact and heuristic approaches

Conclusions  We have witnessed enormous progress in vehicle routing over the past 40 years  We can all take pride in the many successful implementations of vehicle routing software  Still, there is so much more work for us to do 30