Feature-based tuning of simulated annealing applied to the curriculum-based course timetabling problem

We consider the university course timetabling problem, which is one of the most studied problems in educational timetabling. In particular, we focus our attention on the formulation known as the curriculum-based course timetabling problem (CB-CTT), which has been tackled by many researchers and for which there are many available benchmarks.The contribution of this paper is twofold. First, we propose an effective and robust single-stage simulated annealing method for solving the problem. Second, we design and apply an extensive and statistically-principled methodology for the parameter tuning procedure. The outcome of this analysis is a methodology for modeling the relationship between search method parameters and instance features that allows us to set the parameters for unseen instances on the basis of a simple inspection of the instance itself. Using this methodology, our algorithm, despite its apparent simplicity, has been able to achieve high quality results on a set of popular benchmarks.A final contribution of the paper is a novel set of real-world instances, which could be used as a benchmark for future comparison.     

Real-life examination timetabling

An examination timetabling problem at a large American university is presented. Although there are some important differences, the solution approach is based on the ITC 2007 winning solver which is integrated in the open source university timetabling system UniTime. In this work, nine real world benchmark data sets are made publicly available and the results on four of them are presented in this paper. A new approach to further decreasing the number of student conflicts by allowing some exams to be split into multiple examination periods is also studied.     

A graph coloring constructive hyper-heuristic for examination timetabling problems

In this work we investigate a new graph coloring constructive hyper-heuristic for solving examination timetabling problems. We utilize the hierarchical hybridizations of four low level graph coloring heuristics, these being largest degree, saturation degree, largest colored degree and largest enrollment. These are hybridized to produce four ordered lists. For each list, the difficulty index of scheduling the first exam is calculated by considering its order in all lists to obtain a combined evaluation of its difficulty. The most difficult exam to be scheduled is scheduled first (i.e. the one with the minimum difficulty index). To improve the effectiveness of timeslot selection, a?roulette wheel selection mechanism is included in the algorithm to probabilistically select an appropriate timeslot for the chosen exam. We test our proposed approach on the most widely used un-capacitated Carter benchmarks and also on the recently introduced examination timetable dataset from the 2007 International Timetabling Competition. Compared against other methodologies, our results demonstrate that the graph coloring constructive hyper-heuristic produces good results and outperforms other approaches on some of the benchmark instances.     

A new lower bound for curriculum-based course timetabling

In this paper, we propose a new method to compute lower bounds for curriculum-based course timetabling (CTT), which calls for the best weekly assignment of university course lectures to rooms and time slots. The lower bound is obtained by splitting the objective function into two parts, considering one separate problem for each part of the objective function, and summing up the corresponding optimal values (or, in some cases, lower bounds on these values), found by formulating the two parts as Integer Linear Programs (ILPs). The solution of one ILP is obtained by using a column generation procedure. Experimental results show that the proposed lower bound is often better than the ones found by the previous methods in the literature, and also much better than those found by other new ILP formulations illustrated in this paper. The proposed approach is able to obtain improved lower bounds on real-world benchmark instances from the literature, used in the international timetabling competitions ITC2002 and ITC2007, proving for the first time that some of the best-known heuristic solutions are indeed optimal (or close to the optimal ones).     

Variable Neighborhood Search based algorithms for high school timetabling

This work presents the application of Variable Neighborhood Search (VNS) based algorithms to the High School Timetabling Problem. The addressed model of the problem was proposed by the Third International Timetabling Competition (ITC 2011), which released many instances from educational institutions around the world and attracted 17 competitors. Some of the VNS algorithm variants were able to outperform the winner of Third ITC solver, which proposed a Simulated Annealing – Iterated local Search approach. This result coupled with another reports in the literature points that VNS based algorithms are a practical solution method for providing high quality solutions for some hard timetabling problems. Moreover they are easy to implement with few parameters to adjust.     

Design, engineering, and experimental analysis of a simulated annealing approach to the post-enrolment course timetabling problem

The post-enrolment course timetabling (PE-CTT) is one of the most studied timetabling problems, for which many instances and results are available. In this work we design a metaheuristic approach based on simulated annealing to solve the PE-CTT. We consider all the different variants of the problem that have been proposed in the literature and we perform a comprehensive experimental analysis on all the available public instances. The outcome is that our solver, properly engineered and tuned, performs very well on all cases, providing the new best known results on many instances and state-of-the-art values for the others.     

An Iterative local-search framework for solving constraint satisfaction problem

In this article, we introduce a new solving framework based on using alternatively two local-search algorithms to solve constraint satisfaction and optimization problems. The technique presented is based on the integration of local-search algorithm as a mechanism to diversify the search instead of using a build on diversification mechanisms. Thus, we avoid tuning the multiple parameters to escape from a local optimum. This technique improves the existing methods: it is generic especially when the given problem can be expressed as a constraint satisfaction problem. We present the way the local-search algorithm can be used to diversify the search in order to solve real examination timetabling problems. We describe how the local-search algorithm can be used to assist any other specific local-search algorithm to escape from local optimality. We showed that such framework is efficient on real benchmarks for timetabling problems.     

An informed genetic algorithm for the examination timetabling problem

This paper presents the results of a study conducted to investigate the use of genetic algorithms (GAs) as a means of inducing solutions to the examination timetabling problem (ETP). This study differs from previous efforts applying genetic algorithms to this domain in that firstly it takes a two-phased approach to the problem which focuses on producing timetables that meet the hard constraints during the first phase, while improvements are made to these timetables in the second phase so as to reduce the soft constraint costs. Secondly, domain specific knowledge in the form of heuristics is used to guide the evolutionary process. The system was tested on a set of 13 real-world problems, namely, the Carter benchmarks. The performance of the system on the benchmarks is comparable to that of other evolutionary techniques and in some cases the system was found to outperform these techniques. Furthermore, the quality of the examination timetables evolved is within range of the best results produced in the field.     

Modelling and Solving Employee Timetabling Problems

Employee timetabling is the operation of assigning employees to tasks in a set of shifts during a fixed period of time, typically a week. We present a general definition of employee timetabling problems (ETPs) that captures many real-world problem formulations and includes complex constraints. The proposed model of ETPs can be represented in a tabular form that is both intuitive and efficient for constraint representation and processing. The constraint networks of ETPs include non-binary constraints and are difficult to formulate in terms of simple constraint solvers. We investigate the use of local search techniques for solving ETPs. In particular, we propose several versions of hill-climbing that make use of a novel search space that includes also partial assignments. We show that, on large and difficult instances of real world ETPs, where systematic search fails, local search methods perform well and solve the hardest instances. According to our experimental results on various techniques, a simple version of hill climbing based on random moves is the best method for solving large ETP instances.     

A hybrid self-adaptive bees algorithm for examination timetabling problems

A hybrid self-adaptive bees algorithm is proposed for the examination timetabling problems. The bees algorithm (BA) is a population-based algorithm inspired by the way that honey bees forage for food. The algorithm presents a type of neighbourhood search that includes a random search that can be used for optimisation problems. In the BA, the bees search randomly for food sites and return back to the hive carrying the information about the food sites (fitness values); then, other bees will select the sites based on their information (more bees are recruited to the best sites) and will start a random search. We propose three techniques (i.e. disruptive, tournament and rank selection strategies) for selecting the sites, rather than using the fitness value, to improve the diversity of the population. Additionally, a self-adaptive strategy for directing the neighbourhood search is added to further enhance the local intensification capability. Finally, a modified bees algorithm is combined with a local search (i.e. simulated annealing, late acceptance hill climbing) to quickly descend to the optimum solution. Experimental results comparing our proposed modifications with each other and with the basic BA show that all of the modifications outperform the basic BA; an overall comparison has been made with the best known results on two examination timetabling benchmark datasets, which shows that our approach is competitive and works well across all of the problem instances.     

Fairness in examination timetabling: Student preferences and extended formulations

Variations of the examination timetabling problem have been investigated by the research community for more than two decades. The common characteristic between all problems is the fact that the definitions and datasets used all originate from actual educational institutions, particularly universities, including specific examination criteria and the students involved. Although much has been achieved and published on the state-of-the-art problem modelling and optimisation, a lack of attention has been focussed on the students involved in the process. This work presents and utilises the results of an extensive survey seeking student preferences with regard to their individual examination timetables, with the aim of producing solutions which satisfy these preferences while still also satisfying all existing benchmark considerations. The study reveals one of the main concerns relates to fairness within the student's cohort; i.e. a student considers fairness with respect to the examination timetables of their immediate peers, as highly important. Considerations such as providing an equitable distribution of preparation time between all student cohort examinations, not just a majority, are used to form a measure of fairness. In order to satisfy this requirement, we propose an extension to the state-of-the-art examination timetabling problem models widely used in the scientific literature. Fairness is introduced as a new objective in addition to the standard objectives, creating a multi-objective problem. Several real-world examination data models are extended and the benchmarks for each are used in experimentation to determine the effectiveness of a multi-stage multi-objective approach based on weighted Tchebyceff scalarisation in improving fairness along with the other objectives. The results show that the proposed model and methods allow for the production of high quality timetable solutions while also providing a trade-off between the standard soft constraints and a desired fairness for each student.     

Effective local search algorithms for high school timetabling problems

This paper addresses the high school timetabling problem. The problem consists in building weekly timetables for meetings between classes and teachers with the goal of minimizing violations of specific requirements. In the last decades, several mixed-integer programs have been proposed and tested for this family of problems. However, medium and large size instances are still not effectively solved by these programs using state-of-the-art solvers and the scientific community has given special attention to the devising of alternative soft computing algorithms. In this paper, we propose a soft computing approach based on Iterated Local Search and Variable Neighborhood Search metaheuristic frameworks. Our algorithms incorporate new neighborhood structures and local search routines to perform an effective search. We validated the proposed algorithms on variants of the problem using seven public instances and a new dataset with 34 real-world instances including large cases. The results demonstrate that the proposed algorithms outperform the state-of-the-art approaches in both cases, finding the best solutions in 38 out of the 41 tested instances.     

A graph-based MIP formulation of the International Timetabling Competition 2019

Journal of Scheduling - The International Timetabling Competition 2019 (ITC 2019) posed a university timetabling problem involving assigning classes to times and rooms for an entire semester while...     

A parallelized matheuristic for the International Timetabling Competition 2019

Journal of Scheduling - The International Timetabling Competition 2019 (ITC 2019) presents a novel and generalized university timetabling problem composed of traditional class time and room...     

Introducing UniCorT: an iterative university course timetabling tool with MaxSAT

Journal of Scheduling - This paper describes the UniCorT tool designed to solve university course timetabling problems specifically tailored for the 2019 International Timetabling Competition (ITC...     

Case-based heuristic selection for timetabling problems

This paper presents a case-based heuristic selection approach for automated university course and exam timetabling. The method described in this paper is motivated by the goal of developing timetabling systems that are fundamentally more general than the current state of the art. Heuristics that worked well in previous similar situations are memorized in a case base and are retrieved for solving the problem in hand. Knowledge discovery techniques are employed in two distinct scenarios. Firstly, we model the problem and the problem solving situations along with specific heuristics for those problems. Secondly, we refine the case base and discard cases which prove to be non-useful in solving new problems. Experimental results are presented and analyzed. It is shown that case based reasoning can act effectively as an intelligent approach to learn which heuristics work well for particular timetabling situations. We conclude by outlining and discussing potential research issues in this critical area of knowledge discovery for different difficult timetabling problems.     

Integrating employee timetabling with scheduling of machines and transporters in a job-shop environment: A mathematical formulation and an Anarchic Society Optimization algorithm

This paper addresses a ternary-integration scheduling problem that incorporates employee timetabling into the scheduling of machines and transporters in a job-shop environment with a finite number of heterogeneous transporters where the objective is to minimize the completion time of all jobs. The problem is first formulated as a mixed-integer linear programming model. Then, an Anarchic Society Optimization (ASO) algorithm is developed to solve large-sized instances of the problem. The formulation is used to solve small-sized instances and to evaluate the quality of the solutions obtained for instances with larger sizes. A comprehensive numerical study is carried out to assess the performance of the proposed ASO algorithm. The algorithm is compared with three alternative metaheuristic algorithms. It is also compared with several algorithms developed in the literature for the integrated scheduling of machines and transporters. Moreover, the algorithms are tested on a set of adapted benchmark instances for an integrated problem of machine scheduling and employee timetabling. The numerical analysis suggests that the ASO algorithm is both effective and efficient in solving large-sized instances of the proposed integrated job-shop scheduling problem.     

Solving timetabling problems using a cultural algorithm

This paper addresses the solution of timetabling problems using cultural algorithms. The core idea is to extract problem domain information during the evolutionary search, and then combine it with some previously proposed operators, in order to improve performance. The proposed approach is validated using a benchmark of 20 instances, and its results are compared with respect to three other approaches: two evolutionary algorithms and simulated annealing, all of which have been previously adopted to solve timetabling problems.     

Measuring instance difficulty for combinatorial optimization problems

Discovering the conditions under which an optimization algorithm or search heuristic will succeed or fail is critical for understanding the strengths and weaknesses of different algorithms, and for automated algorithm selection. Large scale experimental studies - studying the performance of a variety of optimization algorithms across a large collection of diverse problem instances - provide the resources to derive these conditions. Data mining techniques can be used to learn the relationships between the critical features of the instances and the performance of algorithms. This paper discusses how we can adequately characterize the features of a problem instance that have impact on difficulty in terms of algorithmic performance, and how such features can be defined and measured for various optimization problems. We provide a comprehensive survey of the research field with a focus on six combinatorial optimization problems: assignment, traveling salesman, and knapsack problems, bin-packing, graph coloring, and timetabling. For these problems - which are important abstractions of many real-world problems - we review hardness-revealing features as developed over decades of research, and we discuss the suitability of more problem-independent landscape metrics. We discuss how the features developed for one problem may be transferred to study related problems exhibiting similar structures.     

A Two-Stage Decomposition of High School Timetabling applied to cases in Denmark

Integer Programming (IP) has been used to model educational timetabling problems since the very early days of Operations Research. It is well recognized that these IP models in general are hard to solve, and this area of research is dominated by heuristic solution approaches. In this paper a Two-Stage Decomposition of an IP model for a practical case of high school timetabling is shown. This particular timetabling problem consists of assigning lectures to both a timeslot and a classroom, which is modeled using a very large amount of binary variables. The decomposition splits this model into two separate problems (Stage I and Stage II) with far less variables. These two separate problems are solved in sequence, such that the solution for the Stage I model is given as input to the Stage II model, implying that irreversible decisions are made in Stage I. However, the objective of the Stage II model is partly incorporated in the Stage I model by exploiting that Stage II can be seen as a minimum weight maximum matching problem in a bipartite graph. This theoretically strengthens the decomposition in terms of global optimality. The approach relies on Hall's theorem for the existence of matchings in bipartite graphs, which in its basic form yields an exponential amount of constraints in the Stage I model. However, it is shown that only a small subset of these constraints is needed, making the decomposition tractable in practice for IP solvers. To evaluate the decomposition, 100 real-life problem instances from the database of the high school ERP system Lectio are used. Computational results show that the decomposition performs significantly better than solving the original IP, in terms of both found solutions and bounds.