In this paper we investigate and compare multi-objective and weighted single objective approaches to a real world workforce scheduling problem. For this difficult problem we consider the trade off in solution quality versus population diversity, for different sets of fixed objective weights. Our real-world workforce scheduling problem consists of assigning resources with the appropriate skills to geographically dispersed task locations while satisfying time window constraints. The problem is NP-Hard and contains the Resource Constrained Project Scheduling Problem (RCPSP) as a sub problem. We investigate a genetic algorithm and serial schedule generation scheme together with various multi-objective approaches. We show that multi-objective genetic algorithms can create solutions whose fitness is within 2% of genetic algorithms using weighted sum objectives even though the multi-objective approaches know nothing of the weights. The result is highly significant for complex real-world problems where objective weights are seldom known in advance since it suggests that a multi-objective approach can generate a solution close to the user preferred one without having knowledge of user preferences.

Bien que la gestion des ressources humaines soit une problématique bien étudiée, elle reste d’actualité encore aujourd’hui, notamment en raison de la grande diversité des contextes applicatifs. De plus, les outils d’aide à la décision adressant ces problèmes peuvent encore être améliorés. Dans cette thèse, nous nous intéressons au contexte particulier où les activités des employés correspondent à des tâches fixées dans le temps, requérant des compétences précises et ne pouvant être préemptées. Nous étudions tout d’abord un problème issu de l’industrie pharmaceutique où il s’agit non seulement de trouver une affectation équitable des tâches, mais également de construire les horaires de travail du personnel de manière à respecter les contraintes légales et organisationnelles. Pour résoudre ce problème, nous proposons et comparons deux méthodes exactes, l’une fondée sur un modèle PLNE, l’autre sur un modèle PPC, ainsi que deux méta-heuristiques, l’une reposant sur une décomposition du problème, l’autre reposant sur une recherche par voisinages larges. Nous discutons ensuite de l’intégration de la meilleure de ces méthodes au sein d’un outil d’aide à la décision. Nous nous intéressons ensuite à un problème d’affectation de tâches fixées visant à minimiser le nombre d’employés requis. Pour résoudre ce problème, nous proposons une approche PPC tirant parti de la structure du problème. Nous montrons que cette approche permet d’obtenir rapidement de bonnes bornes, permettant ainsi de prouver l’optimalité sur les instances de la littérature dans un temps raisonnable. / Although workforce scheduling has been studied for decades, it remains highly relevant. In particular, applications and contexts are numerous, and they are always changing. Moreover, decision-support tools dedicated to workforce scheduling could still be improved. In this thesis, we focus on the particular context where workers are assigned to fixed tasks which cannot be preempted and require specific skills. First, we study a problem arising in a company specialized in drug evaluation where fixed tasks have to be assigned in a fair way to qualified employees so that the resulting individual shifts and plannings respect legal and organizational constraints. To handle this problem, we design and compare two exact methods, one based on a MIP model, the other on a CP approach, but also two meta-heuristics, one based on a decomposition strategy, and the other on a large neighborhood search. The best method is integrated within a decision support system. Second, we study a fixed tasks assignment problem, where the objective is to minimize the number of resources required to perform all the tasks. To tackle this problem, we suggest a constraint based approach which takes advantage of the structure of the problem. This approach enables to compute good bounds quickly, and thus, it proves optimality over state-of-the-art instances within a reasonable time limit.

The consumer market of today is characterized by emphasis on superior customer satisfaction and personalization of services. This entails higher customer expectations on organizations, which also includes the workforce scheduling processes in which the consumers expect more decision-power to dictate what they want, when and where they want services to be delivered. For organizations that deliver on-site services, the routing aspect becomes an important part of the scheduling process. Literature on Workforce Scheduling and Routing Problems (WSRP) seldom relate to characteristics of the more dynamic consumer market. As the markets and consumer needs become more flexible, the relevance for research concerning these characteristics increases. This study addresses this by reviewing current literature and present common solution methodologies applied to WSRP, as well as the effects of the online scheduling characteristics. With this as a foundation, a discussion is provided of how WSRP and online scheduling can be combined in order to improve resource utilization and minimize travel time for an on-site service provider. The overall aim of the study is to investigate how an online WSRP with exact time windows can be formulated and solved. The result is a four-stage hybrid method including linear integer programming and constructive heuristics with the objective to minimize travel time, idle time, and the makespan in the schedules. A case study has been conducted on an on-site service provider, and by applying the proposed hybrid methodology on the case company’s scheduling process, results have been obtained that demonstrates improvements of travel time and resource utilization. The study also demonstrate that the appliance of flexible travel times and product dependent service times have positive impact on the quality of the generated schedules. A key insight is that organizations working with exact time windows have to be aware of the trade-off between customer preferences and operational efficiency in day-to-day operations. Thus, organizations have to decide what holds most importance to the organization’s long-term success.

[pt] As atividades de transporte aéreo devem ser fiscalizadas para garantir a adequação dos níveis de segurança e procedimentos operacionais, no Brasil essa atividade é realizada pela Agência Nacional de Aviação Civil (ANAC). Diversos aeroportos devem ser fiscalizados e em cada um deles uma inspeção diferente pode ser necessária. Os inspetores estão alocados em centros da ANAC em diferentes estados e é importante que o custo dessa atividade de inspeção seja minimizado, respeitando as regras existentes. Nesse sentido, essa dissertação de mestrado propõe dois modelos matemáticos para alocação de agentes qualificados para a realização de missões de fiscalização no território brasileiro. O objetivo é a definição de quais colaboradores formarão cada equipe de fiscalização, minimizando o custo de deslocamento dos agentes. O modelo proposto nesse trabalho é multi-período, para o planejamento operacional quinzenal, com a definição do período que as atividades devem ocorrer, da equipe de inspetores multi-habilitados em atividades de inspeção, multi-origens e multi-destinos. A modelagem é feita no LINGO e em Julia com a utilização do pacote JuMP e dos solvers Gurobi e CPLEX. O Modelo 1 propõe uma reformulação de artigos da literatura e possui tempo de solução entre 2 e 25 vezes menor. O Modelo 2 leva em consideração aspectos não considerados até então no Modelo 1, além disso, dada a quantidade de variáveis de decisão, foi utilizada para resolução do modelo heurística baseada na geração de colunas com programação dinâmica, proposta pelo autor, capaz de reduzir em até 95 porcento a quantidade de variáveis de decisão. A heurística permitiu a obtenção de solução inteira em instâncias que não a obtiveram com o modelo completo. / [en] Air transport activities must be inspected to ensure the adequacy of safety levels and operating procedures, in Brazil this activity is carried out by the National Civil Aviation Agency (ANAC). Several airports must be inspected and at each airport a different inspection may be required. The inspectors are located in ANAC centers in different states and it is important that the cost of this inspection activity is minimized, respecting the existing rules. In this sense, this master s thesis proposes two mathematical models for the allocation of qualified workforce to carry out inspection missions in the Brazilian territory. The objective is to define which employees will form in each inspection team, minimizing the cost of displacement of agents. The model proposed in this work is multi-period, for fortnightly operational planning, with the definition of the period that the activities must occur, of a team of multiskilled inspectors in inspection activities, multi-sources and multi-destinations. Modeling is done in LINGO and Julia using the JuMP package and the solvers Gurobi and CPLEX. Model 1 proposes a reformulation of articles in the literature and has a solution time between 2 and 25 times shorter. The second model takes into account aspects not considered so far in Model 1, in addition, given the amount of decision variables, it was used to solve the heuristic model based on the generation of columns with dynamic programming, proposed by the author, capable of reducing by up to 95 percent the amount of decision variables. The heuristic allowed obtaining an integer solution in instances that did not have a solution with the complete model.

Swedish home care industry has been facing both external and internal problems, such as ageing population, varying quality and unsatisfactory continuity. Accordingly, workforce scheduling system, as one of the most common and useful software within home care planning nowadays, is in need of constant improvement and upgrading. This master’s thesis aimed to explore and analyze improvement potential of an established workforce scheduling system for an IT-company. The thesis was divided into two phases, of which a pre-study in Phase I tried to understand the planning process for planners and identify the perceived problems and shortcomings of the current system from a planner’s perspective. Based on the analysis from the pre-study, the caregiver continuity was chosen as the research area for Phase II. The current system was re-implemented and was modelled as an optimization problem. Furthermore, the system mainly consisted of two key parts, mixed integer linear programming (MILP) and heuristics. Different approaches in terms of modifications in both MILP and heuristics were applied to the re-implemented system. The performance of the modifications was measured by multiple evaluation indicators. The test results showed that there was a potential to improve caregiver continuity with 1.2% to almost 13% depending on the modification type. The modifications were lastly suggested for further examination regarding their practical appropriateness by applying them to the current running algorithm. / Den svenska hemtjänsten möter både yttre och inre problem såsom åldrande befolkning, varierande kvalitet och bristande kontinuitet. Schemaläggningssystemet som är en av de vanligaste och användbaraste programvarorna inom hemtjänsten behöver därmed en ständig förbättring och uppgradering som bemöter de existerande utmaningarna. Detta examensarbete hade som syfte att utforska och analysera förbättringspotentialen av ett etablerat schemaläggningssystem för ett ITföretag. Arbetet var indelat i två faser, varav förstudien i Fas I försökte förstå planerarnas planeringsprocesser och identifiera upplevda problem och brister i det nuvarande systemet utifrån ett planerares perspektiv. Baserat på analysen från förstudien, personalkontinuitet valdes som ett forskningsområde för Fas II. Nuvarande systemet implementerades om och det modellerades som ett optimeringsproblem. Systemet bestod huvudsakligen av två nyckeldelar, blandat heltalslinjärprogrammering (MILP) och heuristik. Olika metoder i form av modifieringar i både MILP och heuristik tillämpades på det omimplementerade systemet. Modifieringarnas prestanda mättes sedan med flera utvärderingsindikatorer. Testresultaten visade att, beroende på vilken modifiering det gäller, fanns det en potential att förbättra personalkontinuiteten med 1,2% till nästan 13%. Det föreslogs slutligen att modifieringarnas praktiska lämplighet behövs undersökas ytterligare genom att applicera det på det nuvarande systemet som är i drift.