'The application of Bayesian Optimization and Classifier Systems in Nurse Scheduling'

Abstract. Two ideas taken from Bayesian optimization and classifier systems are presented for personnel scheduling based on choosing a suitable scheduling rule from a set for each person's assignment. Unlike our previous work of using genetic algorithms whose learning is implicit, the learning in both approaches is explicit, i.e. we are able to identify building blocks directly. To achieve this target, the Bayesian optimization algorithm builds a Bayesian network of the joint probability distribution of the rules used to construct solutions, while the adapted classifier system assigns each rule a strength value that is constantly updated according to its usefulness in the current situation. Computational results from 52 real data instances of nurse scheduling demonstrate the success of both approaches. It is also suggested that the learning mechanism in the proposed approaches might be suitable for other scheduling problems.

'Bayesian Optimisation for Nurse Scheduling'

A Bayesian optimisation algorithm for a nurse scheduling problem is presented, which involves choosing a suitable scheduling rule from a set for each nurse's assignment. When a human scheduler works, he normally builds a schedule systematically following a set of rules. After much practice, the scheduler gradually masters the knowledge of which solution parts go well with others. He can identify good parts and is aware of the solution quality even if the scheduling process is not yet completed, thus having the ability to finish a schedule by using flexible, rather than fixed, rules. In this paper, we design a more human-like scheduling algorithm, by using a Bayesian optimisation algorithm to implement explicit learning from past solutions. A nurse scheduling problem from a UK hospital is used for testing. Unlike our previous work that used Genetic Algorithms to implement implicit learning [1], the learning in the proposed algorithm is explicit, i.e. we identify and mix building blocks directly. The Bayesian optimisation algorithm is applied to implement such explicit learning by building a Bayesian network of the joint distribution of solutions. The conditional probability of each variable in the network is computed according to an initial set of promising solutions. Subsequently, each new instance for each variable is generated by using the corresponding conditional probabilities, until all variables have been generated, i.e. in our case, new rule strings have been obtained. Sets of rule strings are generated in this way, some of which will replace previous strings based on fitness. If stopping conditions are not met, the conditional probabilities for all nodes in the Bayesian network are updated again using the current set of promising rule strings. For clarity, consider the following toy example of scheduling five nurses with two rules (1: random allocation, 2: allocate nurse to low-cost shifts). In the beginning of the search, the probabilities of choosing rule 1 or 2 for each nurse is equal, i.e. 50%. After a few iterations, due to the selection pressure and reinforcement learning, we experience two solution pathways: Because pure low-cost or random allocation produces low quality solutions, either rule 1 is used for the first 2-3 nurses and rule 2 on remainder or vice versa. In essence, Bayesian network learns 'use rule 2 after 2-3x using rule 1' or vice versa. It should be noted that for our and most other scheduling problems, the structure of the network model is known and all variables are fully observed. In this case, the goal of learning is to find the rule values that maximize the likelihood of the training data. Thus, learning can amount to 'counting' in the case of multinomial distributions. For our problem, we use our rules: Random, Cheapest Cost, Best Cover and Balance of Cost and Cover. In more detail, the steps of our Bayesian optimisation algorithm for nurse scheduling are: 1. Set t = 0, and generate an initial population P(0) at random; 2. Use roulette-wheel selection to choose a set of promising rule strings S(t) from P(t); 3. Compute conditional probabilities of each node according to this set of promising solutions; 4. Assign each nurse using roulette-wheel selection based on the rules' conditional probabilities. A set of new rule strings O(t) will be generated in this way; 5. Create a new population P(t+1) by replacing some rule strings from P(t) with O(t), and set t = t+1; 6. If the termination conditions are not met (we use 2000 generations), go to step 2. Computational results from 52 real data instances demonstrate the success of this approach. They also suggest that the learning mechanism in the proposed approach might be suitable for other scheduling problems. Another direction for further research is to see if there is a good constructing sequence for individual data instances, given a fixed nurse scheduling order. If so, the good patterns could be recognized and then extracted as new domain knowledge. Thus, by using this extracted knowledge, we can assign specific rules to the corresponding nurses beforehand, and only schedule the remaining nurses with all available rules, making it possible to reduce the solution space. Acknowledgements The work was funded by the UK Government's major funding agency, Engineering and Physical Sciences Research Council (EPSRC), under grand GR/R92899/01. References [1] Aickelin U, "An Indirect Genetic Algorithm for Set Covering Problems", Journal of the Operational Research Society, 53(10): 1118-1126,

'A Bayesian Optimisation Algorithm for the Nurse Scheduling Problem'

Abstract- A Bayesian optimization algorithm for the nurse scheduling problem is presented, which involves choosing a suitable scheduling rule from a set for each nurse's assignment. Unlike our previous work that used GAs to implement implicit learning, the learning in the proposed algorithm is explicit, i.e. eventually, we will be able to identify and mix building blocks directly. The Bayesian optimization algorithm is applied to implement such explicit learning by building a Bayesian network of the joint distribution of solutions. The conditional probability of each variable in the network is computed according to an initial set of promising solutions. Subsequently, each new instance for each variable is generated by using the corresponding conditional probabilities, until all variables have been generated, i.e. in our case, a new rule string has been obtained. Another set of rule strings will be generated in this way, some of which will replace previous strings based on fitness selection. If stopping conditions are not met, the conditional probabilities for all nodes in the Bayesian network are updated again using the current set of promising rule strings. Computational results from 52 real data instances demonstrate the success of this approach. It is also suggested that the learning mechanism in the proposed approach might be suitable for other scheduling problems.

'Explicit Learning: an Effort towards Human Scheduling Algorithms'

Abstract Scheduling problems are generally NP-hard combinatorial problems, and a lot of research has been done to solve these problems heuristically. However, most of the previous approaches are problem-specific and research into the development of a general scheduling algorithm is still in its infancy. Mimicking the natural evolutionary process of the survival of the fittest, Genetic Algorithms (GAs) have attracted much attention in solving difficult scheduling problems in recent years. Some obstacles exist when using GAs: there is no canonical mechanism to deal with constraints, which are commonly met in most real-world scheduling problems, and small changes to a solution are difficult. To overcome both difficulties, indirect approaches have been presented (in [1] and [2]) for nurse scheduling and driver scheduling, where GAs are used by mapping the solution space, and separate decoding routines then build solutions to the original problem. In our previous indirect GAs, learning is implicit and is restricted to the efficient adjustment of weights for a set of rules that are used to construct schedules. The major limitation of those approaches is that they learn in a non-human way: like most existing construction algorithms, once the best weight combination is found, the rules used in the construction process are fixed at each iteration. However, normally a long sequence of moves is needed to construct a schedule and using fixed rules at each move is thus unreasonable and not coherent with human learning processes. When a human scheduler is working, he normally builds a schedule step by step following a set of rules. After much practice, the scheduler gradually masters the knowledge of which solution parts go well with others. He can identify good parts and is aware of the solution quality even if the scheduling process is not completed yet, thus having the ability to finish a schedule by using flexible, rather than fixed, rules. In this research we intend to design more human-like scheduling algorithms, by using ideas derived from Bayesian Optimization Algorithms (BOA) and Learning Classifier Systems (LCS) to implement explicit learning from past solutions. BOA can be applied to learn to identify good partial solutions and to complete them by building a Bayesian network of the joint distribution of solutions [3]. A Bayesian network is a directed acyclic graph with each node corresponding to one variable, and each variable corresponding to individual rule by which a schedule will be constructed step by step. The conditional probabilities are computed according to an initial set of promising solutions. Subsequently, each new instance for each node is generated by using the corresponding conditional probabilities, until values for all nodes have been generated. Another set of rule strings will be generated in this way, some of which will replace previous strings based on fitness selection. If stopping conditions are not met, the Bayesian network is updated again using the current set of good rule strings. The algorithm thereby tries to explicitly identify and mix promising building blocks. It should be noted that for most scheduling problems the structure of the network model is known and all the variables are fully observed. In this case, the goal of learning is to find the rule values that maximize the likelihood of the training data. Thus learning can amount to 'counting' in the case of multinomial distributions. In the LCS approach, each rule has its strength showing its current usefulness in the system, and this strength is constantly assessed [4]. To implement sophisticated learning based on previous solutions, an improved LCS-based algorithm is designed, which consists of the following three steps. The initialization step is to assign each rule at each stage a constant initial strength. Then rules are selected by using the Roulette Wheel strategy. The next step is to reinforce the strengths of the rules used in the previous solution, keeping the strength of unused rules unchanged. The selection step is to select fitter rules for the next generation. It is envisaged that the LCS part of the algorithm will be used as a hill climber to the BOA algorithm. This is exciting and ambitious research, which might provide the stepping-stone for a new class of scheduling algorithms. Data sets from nurse scheduling and mall problems will be used as test-beds. It is envisaged that once the concept has been proven successful, it will be implemented into general scheduling algorithms. It is also hoped that this research will give some preliminary answers about how to include human-like learning into scheduling algorithms and may therefore be of interest to researchers and practitioners in areas of scheduling and evolutionary computation. References 1. Aickelin, U. and Dowsland, K. (2003) 'Indirect Genetic Algorithm for a Nurse Scheduling Problem', Computer & Operational Research (in print). 2. Li, J. and Kwan, R.S.K. (2003), 'Fuzzy Genetic Algorithm for Driver Scheduling', European Journal of Operational Research 147(2): 334-344. 3. Pelikan, M., Goldberg, D. and Cantu-Paz, E. (1999) 'BOA: The Bayesian Optimization Algorithm', IlliGAL Report No 99003, University of Illinois. 4. Wilson, S. (1994) 'ZCS: A Zeroth-level Classifier System', Evolutionary Computation 2(1), pp 1-18.

An Estimation of Distribution Algorithm for Nurse Scheduling

Schedules can be built in a similar way to a human scheduler by using a set of rules that involve domain knowledge. This paper presents an Estimation of Distribution Algorithm (EDA) for the nurse scheduling problem, which involves choosing a suitable scheduling rule from a set for the assignment of each nurse. Unlike previous work that used Genetic Algorithms (GAs) to implement implicit learning, the learning in the proposed algorithm is explicit, i.e. we identify and mix building blocks directly. The EDA is applied to implement such explicit learning by building a Bayesian network of the joint distribution of solutions. The conditional probability of each variable in the network is computed according to an initial set of promising solutions. Subsequently, each new instance for each variable is generated by using the corresponding conditional probabilities, until all variables have been generated, i.e. in our case, a new rule string has been obtained. Another set of rule strings will be generated in this way, some of which will replace previous strings based on fitness selection. If stopping conditions are not met, the conditional probabilities for all nodes in the Bayesian network are updated again using the current set of promising rule strings. Computational results from 52 real data instances demonstrate the success of this approach. It is also suggested that the learning mechanism in the proposed approach might be suitable for other scheduling problems.

An evolutionary squeaky wheel optimisation approach to personnel scheduling

The quest for robust heuristics that are able to solve more than one problem is ongoing. In this paper, we present, discuss and analyse a technique called Evolutionary Squeaky Wheel Optimisation and apply it to two different personnel scheduling problems. Evolutionary Squeaky Wheel Optimisation improves the original Squeaky Wheel Optimisation’s effectiveness and execution speed by incorporating two additional steps (Selection and Mutation) for added evolution. In the Evolutionary Squeaky Wheel Optimisation, a cycle of Analysis-Selection-Mutation-Prioritization-Construction continues until stopping conditions are reached. The aim of the Analysis step is to identify below average solution components by calculating a fitness value for all components. The Selection step then chooses amongst these underperformers and discards some probabilistically based on fitness. The Mutation step further discards a few components at random. Solutions can become incomplete and thus repairs may be required. The repair is carried out by using the Prioritization step to first produce priorities that determine an order by which the following Construction step then schedules the remaining components. Therefore, improvements in the Evolutionary Squeaky Wheel Optimisation is achieved by selective solution disruption mixed with iterative improvement and constructive repair. Strong experimental results are reported on two different domains of personnel scheduling: bus and rail driver scheduling and hospital nurse scheduling.

A Bayesian optimization algorithm for the nurse scheduling problem

A Bayesian optimization algorithm for the nurse scheduling problem is presented, which involves choosing a suitable scheduling rule from a set for each nurse’s assignment. Unlike our previous work that used GAs to implement implicit learning, the learning in the proposed algorithm is explicit, i.e. eventually, we will be able to identify and mix building blocks directly. The Bayesian optimization algorithm is applied to implement such explicit learning by building a Bayesian network of the joint distribution of solutions. The conditional probability of each variable in the network is computed according to an initial set of promising solutions. Subsequently, each new instance for each variable is generated by using the corresponding conditional probabilities, until all variables have been generated, i.e. in our case, a new rule string has been obtained. Another set of rule strings will be generated in this way, some of which will replace previous strings based on fitness selection. If stopping conditions are not met, the conditional probabilities for all nodes in the Bayesian network are updated again using the current set of promising rule strings. Computational results from 52 real data instances demonstrate the success of this approach. It is also suggested that the learning mechanism in the proposed approach might be suitable for other scheduling problems.

BOA for nurse scheduling

Our research has shown that schedules can be built mimicking a human scheduler by using a set of rules that involve domain knowledge. This chapter presents a Bayesian Optimization Algorithm (BOA)for the nurse scheduling problem that chooses such suitable scheduling rules from a set for each nurse’s assignment. Based on the idea of using probabilistic models, the BOA builds a Bayesian network for the set of promising solutions and samples these networks to generate new candidate solutions. Computational results from 52 real data instances demonstrate the success of this approach. It is also suggested that the learning mechanism in the proposed algorithm may be suitable for other scheduling problems.

Some heuristic algorithms for total tardiness minimization in a flowshop with blocking

The flowshop scheduling problem with blocking in-process is addressed in this paper. In this environment, there are no buffers between successive machines: therefore intermediate queues of jobs waiting in the system for their next operations are not allowed. Heuristic approaches are proposed to minimize the total tardiness criterion. A constructive heuristic that explores specific characteristics of the problem is presented. Moreover, a GRASP-based heuristic is proposed and Coupled with a path relinking strategy to search for better outcomes. Computational tests are presented and the comparisons made with an adaptation of the NEH algorithm and with a branch-and-bound algorithm indicate that the new approaches are promising. (c) 2007 Elsevier Ltd. All rights reserved.

BASS 4: a software system for ergonomic design and evaluation of working hours

OBJECTIVE: To extend an existing computer programme for the evaluation and design of shift schedules (BASS 3) by integrating workload as well as economic aspects. METHODS: The redesigned prototype BASS 4 includes a new module with a suitable and easily applicable screening method (EBA) for the assessment of the intensity of physical, emotional and cognitive workload components and their temporal patterns. Specified criterion functions based on these ratings allow for an adjustment of shift and rest duration according to the intensity of physical and mental workload. Furthermore, with regard to interactive effects both workload and temporal conditions, e.g. time of day, are taken into account. In a second new module, important economic aspects and criteria have been implemented. Different ergonomic solutions for scheduling problems can now also be evaluated with regard to their economic costs. RESULTS: The new version of the computer programme (BASS 4) can now simultaneously take into account numerous ergonomic, legal, agreed and economic criteria for the design and evaluation of working hours. CONCLUSIONS: BASS 4 can now be used as an instrument for the design and the evaluation of working hours with regard to legal, ergonomic and economic aspects at the shop floor as well as in administrative (e.g. health and safety inspection) and research problems.

Análise de risco em sequenciamento de actividades com restrições de recursos

Mestrado em Engenharia Electrotécnica e de Computadores. Área de Especialização em Sistemas e Planeamento Industrial.

Aspectos de aprendizagem em optimização

A optimização e a aprendizagem em Sistemas Multi-Agente são consideradas duas áreas promissoras mas relativamente pouco exploradas. A optimização nestes ambientes deve ser capaz de lidar com o dinamismo. Os agentes podem alterar o seu comportamento baseando-se em aprendizagem recente ou em objectivos de optimização. As estratégias de aprendizagem podem melhorar o desempenho do sistema, dotando os agentes da capacidade de aprender, por exemplo, qual a técnica de optimização é mais adequada para a resolução de uma classe particular de problemas, ou qual a parametrização é mais adequada em determinado cenário. Nesta dissertação são estudadas algumas técnicas de resolução de problemas de Optimização Combinatória, sobretudo as Meta-heurísticas, e é efectuada uma revisão do estado da arte de Aprendizagem em Sistemas Multi-Agente. É também proposto um módulo de aprendizagem para a resolução de novos problemas de escalonamento, com base em experiência anterior. O módulo de Auto-Optimização desenvolvido, inspirado na Computação Autónoma, permite ao sistema a selecção automática da Meta-heurística a usar no processo de optimização, assim como a respectiva parametrização. Para tal, recorreu-se à utilização de Raciocínio baseado em Casos de modo que o sistema resultante seja capaz de aprender com a experiência adquirida na resolução de problemas similares. Dos resultados obtidos é possível concluir da vantagem da sua utilização e respectiva capacidade de adaptação a novos e eventuais cenários.

Problema de atribuição de vigilâncias de exames a docentes

O presente trabalho foi realizado com o intuito de resolver o problema de alocação de vigilantes a exames do Instituto Superior de Engenharia do Porto, no departamento de Engenharia Mecânica. O modelo apresentado faz a atribuição das vigilâncias de uma forma hierárquica, utilizando vários critérios, desde a regência da unidade curricular até à simples vigilância. Devido ao facto de estar implementado informaticamente, apresenta reduzidos tempos na formulação e obtenção de uma solução, o que o torna uma boa ferramenta para a criação de cenários alternativos. Em suma, o modelo proposto neste trabalho apresenta soluções de melhor qualidade, em que a distribuição de afetações é proporcional entre os docentes, e o seu tempo de obtenção é muito reduzido em comparação com a alternativa atual.

Afetação de recursos em sistemas de produção

O problema do escalonamento, por ser um dos factores fundamentais na tomada de decisão para uma boa gestão das operações, tem sido alvo de um amplo estudo, tanto na sua componente teórica como na sua componente prática. A importância de um escalonamento correto das operações é preponderante, quando as pequenas diferenças, em termos de tempos de produção, podem ter um grande impacto na competitividade da organização. Em muitas unidades produtivas, existem máquinas capazes de realizar as mesmas operações com diferentes desempenhos. Isto pode dever-se à necessidade de flexibilizar os recursos ou mesmo a uma atualização da capacidade produtiva. Embora os problemas de máquinas diferentes em paralelo tenham sido alvo de um vasto estudo, muitos deles não são passíveis de ser resolvidos através de métodos exatos. O problema de minimização do makespan (Rm||Cmax), é NP-hard, sendo habitualmente abordado através de heurísticas. Entre as heurísticas utilizadas em problemas de minimização do makespan em máquinas diferentes em paralelo, é possível identificar duas filosofias de afectação: a que utiliza os tempos de processamento para alocar as tarefas e a que utiliza as datas de conclusão. Nesta dissertação, pretende-se dar uma contribuição para a resolução do problema de afectação de recursos em sistemas de produção. Para tal, foram propostas as heurísticas OMTC 3 e Suffrage One. A contribuição consiste na proposta de versões híbridas e modificadas das heurística MCT e Suffrage, uma vez identificadas várias características que podem limitar o seu desempenho, como o facto da heurística MCT alocar as tarefas numa ordem aleatória ou a heurística Suffrage alocar mais que uma tarefa por iteração. Finalmente, procedeu-se à realização de testes computacionais, para avaliar o desempenho das heurísticas propostas. Os testes realizados permitiram concluir que a heurística OMTC 3 apresentou um melhor desempenho que a heurística MCT.

Resolução do problema de atribuição de salas a exames

Este trabalho pretende resolver o problema das alocações de salas a exames no Departamento de Engenharia Mecânica do Instituto Superior de Engenharia do Porto. A solução desenvolvida atribui salas a exames respeitando as restrições de capacidade de salas e a restrição de realização dum único exame por sala num determinado período, por forma a minimizar a atribuição de salas e, consequentemente, docentes a exames. Foi criado um modelo matemático, que representa as variáveis relevantes do problema, e realiza a sua implementação numa plataforma informática amigável para o utilizador. O modelo matemático foi validado comparando as suas soluções com as obtidas através do processo manual. Os resultados do novo método demonstram a sua supremacia relativamente ao modelo atual. No futuro, poderá ser estudada a possibilidade de usar esta ferramenta na resolução do mesmo problema em realidades diferentes da do Departamento de Engenharia Mecânica do ISEP.