Atle Riise

Followers

Following

Public Views

Interests

Uploads

Papers by Atle Riise

Scheduling Vehicles with Spatial Conflicts

Transportation Science

When scheduling the movement of individual vehicles on a traffic network, one must ensure that th... more When scheduling the movement of individual vehicles on a traffic network, one must ensure that they never get too close to one another. This is normally modelled by segmenting the network and forbidding two vehicles to occupy the same segment at the same time. This approximation is often insufficient or too restraining. This study develops and systematises the use of conflict regions to model spatial proximity constraints. By extending the classical disjunctive programming approach to job-shop scheduling problems, we demonstrate how conflict regions can be exploited to efficiently schedule the collective movements of a set of vehicles, in this case aircraft moving on an airport ground network. We also show how conflict regions can be used in the short-term control of vehicle speeds to avoid collisions and deadlocks. The overall approach was implemented in a software system for air traffic management at airports and successfully evaluated for scheduling and guiding airplanes during a...

Integrated planning and scheduling in operational patient management

This thesis addresses a class of particularly challenging real world optimisation problems that o... more This thesis addresses a class of particularly challenging real world optimisation problems that occur in operational patient management in hospitals. These problems are found in surgery scheduling, as well as in scheduling for various nonsurgical treatment or diagnostic services. As a class, these problems can be identified by the following properties: They concern the scheduling of a (potentially large) set of patients, where one or several activities must be scheduled for each patient. A set of resources is required for each activity, and there are many alternative such sets to choose from (the problems are multi-modal). Furthermore, the problems involve both planning and scheduling decisions; both a day and an exact start time must also be chosen for each activity. These problems are subject to resource capacity constraints and a variety of time constraints, often including both minimum and maximum time lags, as well as time windows. Common objective functions involve both the quality of treatment and an efficient use of hospital resources. In general, these problems have a high computational complexity, and in many cases even simpler subproblems have been shown to be NP-hard. in the process. I would like to thank my main supervisor, Professor Edmund K. Burke for offering to take me on, and for his insights and contributions throughout these years. I would also like to thank Professor Carlo Mannino, who entered the roles of co-supervisor, colleague, and friend in the latter part of the process. I have learned a lot from our discussions, and hope to continue to do so. A big thanks also to my colleagues in the Department of Applied Mathematics, and especially the good people of the Optimisation group, for maintaining such a friendly and constructive working environment. A special thanks to Oddvar Kloster and Leonardo Lamorgese, with whom I've worked closely in the AKTIV project and had many interesting discussions. Also thanks to Geir Hasle, who employed me at SINTEF all those years ago, and whose input has always been appreciated. Also, I would like to thank my friends for being just that; I'm looking forward to more frequent interactions. Special thanks to Kristin and Arild, who beat me to it. They have been offering encouragement as well as practical help throughout. Finally, I would like to thank my family. Most of all, my wonderful girlfriend and cohabitant Kristin, whose patience and loving support has been indispensable to me. I deeply value our relationship. That post-thesis life starts now! Also, thanks to my daughter, Tanvi. A child when all this began, she has always been a wonderful person and a joy in my life. And to Rufus the dog, for all those long days at the office. Any and all mistakes in this thesis are his.

Download

Stochastic optimization Stokastisk optimering

FILE CODE DATE APPROVED BY (NAME, POSITION, SIGN.) 2005-06-13 Tore Gimse, Research Director Vehic... more FILE CODE DATE APPROVED BY (NAME, POSITION, SIGN.) 2005-06-13 Tore Gimse, Research Director Vehicle routing is a complex optimization problem with numerous variants related to various forms of transport planning. Particularly in an operational planning situation, the dynamic and stochastic aspects of the problems are important. The research project Dynamic Optimization in Transportation (DOiT) addresses vehicle routing in such a dynamic setting. In this survey, we give an overview of some of the literature on dynamic and stochastic vehicle routing problems that we find to be particularly relevant to our research in this field. The survey covers dynamics and stochastics both in travel times and transport requests. Our focus is on algorithms for real world applications, therefore mainly work concerning local search and other heuristic methods are covered. The main body of the paper describes and evaluates modeling and solution approaches published in recent years.

MVA Report Routing Applications in Newspaper Delivery

The goal of this report is to give an up-to-date account of routing applications in the newspaper... more The goal of this report is to give an up-to-date account of routing applications in the newspaper business. We describe the newspaper supply chain, and focus on the “last mile” distribution that has been advocated as an application of arc routing in the literature. A literature survey is provided, followed by a discussion of the arc routing model and its adequacy to newspaper applications. A more general and normally more adequate model: The Node, Edge, and Arc Routing Problem, is discussed. Characteristics of routing problems in carrier delivery are presented, together with a case study from the development of a web-based route design and revision system. Finally, summary, conclusions, and prospects for the future are given. PREPARED BY Geir Hasle SIGNATURE CHECKED BY Atle Riise SIGNATURE APPROVED BY Tomas Nordlander SIGNATURE REPORT NO. A23753 ISBN 978-82-14-05310-4 CLASSIFICATION Unrestricted CLASSIFICATION THIS PAGE Unrestricted

Download

The Surgery Scheduling Problem - A General Model

-The term surgery scheduling is used about a variety of strategic, tactical and operational sched... more -The term surgery scheduling is used about a variety of strategic, tactical and operational scheduling problems, many of which are critical to an efficient use of hospital resources. Our focus is on operational surgery scheduling problems, which are often NP-hard. The exact problem formulation varies substantially among hospitals, or even hospital departments. In addition, the level of detail vary between different planning situations, ranging from long term patient admission planning to a very detailed planning of the same day's surgeries. This diversity makes it difficult to design scheduling methods and software solutions that are applicable to a wide range of surgery scheduling problems, without extensive customization for each individual application. We approach this challenge by proposing a new generalised model for surgery scheduling problems. The problem can be seen as a rich extension to the resource-constrained project scheduling problem, and we present a structured overview of how our contribution relates to the existing project scheduling literature. We represent this problem by extending the classical disjunctive graph model developed for jobshop scheduling problems. To investigate the power of exact optimization methods in solving generalised surgery scheduling problems, we formulate this disjunctive model as a Mixed Integer Linear Program and solve it by means of a commercial solver. The results show that while it is not capable of solving realistic instances to optimality, the formulation produces good bounds, and promising results were found for interesting sub problem

Chapter 3 DYNAMIC AND STOCHASTIC VEHICLE ROUTING IN PRACTICE

The VRP is a key to efficient transportation logistics. It is a computationally very hard problem... more The VRP is a key to efficient transportation logistics. It is a computationally very hard problem. Whereas classical OR models are static and deterministic, these assumptions are rarely warranted in an industrial setting. Lately, there has been an increased focus on dynamic and stochastic vehicle routing in the research community. However, very few generic routing tools based on stochastic or dynamic models are available. We illustrate the need for dynamics and stochastic models in industrial routing, describe the Dynamic and Stochastic VRP, and how we have extended a generic VRP solver to cope with dynamics and uncertainty.

Download

Recursive logic-based Benders’ decomposition for multi-mode outpatient scheduling

European Journal of Operational Research, 2016

Efficient outpatient scheduling is becoming increasingly important for the overall cost effective... more Efficient outpatient scheduling is becoming increasingly important for the overall cost effectiveness and treatment efficiency of a hospital. We consider a class of multi-mode appointment scheduling problems, with variable resource availability and resource setup times. These problems are frequently found in hospital outpatient clinics, and they are typically hard to solve.

Modelling and solving generalised operational surgery scheduling problems

Computers & Operations Research, 2016

Abstract The term ‘surgery scheduling’ is used to describe a variety of strategic, tactical and o... more Abstract The term ‘surgery scheduling’ is used to describe a variety of strategic, tactical and operational scheduling problems, many of which are critical to the quality of treatment and to the efficient use of hospital resources. We consider operational surgery scheduling problems. The exact problem formulation varies substantially between hospitals or, even, hospital departments. In addition, the level of detail varies between different planning situations, ranging from long term patient admission planning to the very detailed scheduling of a particular day׳s surgeries. This diversity makes it difficult to design general scheduling methods and software solutions that can be applied without extensive customisation for each application. We approach this challenge by proposing a new generalised model for surgery scheduling problems. We show how this model extends the multi-project, multi-mode resource constrained project scheduling problem with generalised time constraints, including some extensions that to our knowledge have not been previously studied. Furthermore, we present a search method for solving the proposed model. The algorithm uses on-line learning to balance computational loads between a construction and an improvement method, both working on high level solution representations. An adapted schedule generation scheme is used to map these to concrete schedules. We perform computational experiments using realistic problem instances from three surgery scheduling planning situations at a medium sized Norwegian hospital; day scheduling, week scheduling and admission planning. The results show that the algorithm performs well across these quite different problems without any off-line customisation or parameter tuning.

A mathematical model for the nurse rostering system

The model we present in this report is mainly developed on work done together with the company Ga... more The model we present in this report is mainly developed on work done together with the company Gatsoft AS. With minor changes and extensions, a version of the model has been implemented and delivered to Gatsoft as a plug-in to the rostering part of their software GAT. The model consists of two types of constraints; hard and soft. The hard constraints must never be violated. The origin of the hard constraint is typically legislation or working acts. The soft constraints can be violated, but we try to avoid it. Violating the soft constraints is usually necessary to be able to find a solution. A solution to the model is one in which all hard constraints are satisfied. The quality (objective function value) of the solution is measured in terms of violations to the soft constraints.

Handbook of Healthcare Delivery Systems, chap 29. Scheduling and Sequencing

Creating schedules and sequences is a task which occurs frequently in healthcare environments. So... more Creating schedules and sequences is a task which occurs frequently in healthcare environments. Some of the more common scheduling scenarios include:  Creating rosters for nurse and physicians.  Scheduling patient appointments.  Scheduling the use of operating theatres.  Scheduling regular treatment. For example, radiotherapy and chemotherapy. Often the schedules are still created by hand. Producing them manually without the help of a decision support system has two significant disadvantages: 1. The schedules produced are sub-optimal and do not utilise precious resources as efficiently as possible. This can have a number of adverse effects such as unnecessary expenditure, inferior quality of patient care and dissatisfied employees. 2. Often these time consuming and sometimes frustrating chores fall to employees who could use their time more productively, especially as they are frequently already overloaded. For example, it is usually a responsibility of head nurses to create rost...

The Practice and Theory of Automated Timetabling (2012)

Annals of Operations Research, 2015

This special volume is focused upon a selection of the papers that were presented at the 9th Inte... more This special volume is focused upon a selection of the papers that were presented at the 9th International Conference on the Practice and Theory of Automated Timetabling (PATAT) in Son, Norway, between 28th and 31st August, 2012. The PATAT conferences are held biennially. This is the third occasion that we have employed the Annals of Operations Research as the forum for a revised and selected collection of papers. The first and second PATAT special volumes of this journal (Volumes 194 and 218) represented the 7th and 8th conferences, held in Montreal in 2008 and in Belfast in 2010, respectively. The PATAT series has been running since 1995. It has been held in Scotland, Canada

Download

Dynamic And Stochastic Vehicle Routing In Practice

Operations Research/Computer Science Interfaces Series

A Hybrid Approach for Solving Real-World Nurse Rostering Problems

Lecture Notes in Computer Science, 2011

Nurse rostering is the process of creating a plan for nurse working hours over a given time horiz... more Nurse rostering is the process of creating a plan for nurse working hours over a given time horizon. This problem, most variants of which are NP-hard, has been studied extensively for many years. Still, practical nurse rostering is mostly done manually, often by highly qualified health care personnel. This underlines the need to address the challenges of realistic, applied nurse rostering, and the implementation of advanced rostering methods in commercial software. In this paper, we present an industrial case study of a nurse rostering software currently used in several hospitals and other health care institutions in Norway and Sweden. The presented problem model has a rich set of hard and soft constraints, as required by Norwegian hospitals. Our solution approach is a hybrid: An Iterated Local Search framework that uses Constraint Programming for initial solution construction and diversification, and a Variable Neighborhood Descent for iterative improvement. The search method shows good results in terms of solution quality and computation time on a set of real world instances. We make these test instances available on-line.

Download

A Hybrid Evolutionary Algorithm for the Generalized Surgery Scheduling Problem

Download

Patat 2012

Scheduling and Sequencing

A heuristic approach for solving real world nurse rostering

... Tel.: +47 22 06 73 00 Fax: +47 22 06 73 50 E-mail: helle.froyseth@sintef.no, martin.stolevik@... more

Scalable and instance robust metaheuristics for a rich VRP model

Using Heterogeneous Computing for Solving Vehicle Routing Problems

On parallel local search for permutations

Journal of the Operational Research Society, 2014

We investigate some ways in which massively parallel computing devices can be exploited in local ... more We investigate some ways in which massively parallel computing devices can be exploited in local search algorithms. We show that the substantial speed-ups that can be gained from parallel neighbourhood evaluation enables an efficient best improvement local search, and that this in turn enables further speed-ups through selection and parallel application of a set of independent, improving moves. Our experiments demonstrate a total speed-up of up to several hundred times compared to a classical, sequential best improvement search. We also demonstrate how an exchange of good partial solutions between the incumbent and best found solutions improves the efficiency of the Iterated Local Search algorithm.

Download

Atle Riise

Uploads

Papers by Atle Riise

Log In