Submitted by Silvana Teresinha Dornelles Studzinski (sstudzinski) on 2015-11-04T15:30:21Z No. of bitstreams: 1 Kelly Cristina Ferreira Dias.pdf: 6056954 bytes, checksum: f38a4fe2aa64b584f392432c93244c7f (MD5) / Made available in DSpace on 2015-11-04T15:30:21Z (GMT). No. of bitstreams: 1 Kelly Cristina Ferreira Dias.pdf: 6056954 bytes, checksum: f38a4fe2aa64b584f392432c93244c7f (MD5) Previous issue date: 2015-07-29 / Nenhuma / O desempenho operacional das instituições hospitalares tem relação direta com a escala de trabalho das pessoas. Assim, a tarefa de construção das mesmas exige atenção a diversos detalhes para obtenção de bons resultados. É neste contexto que um modelo, computacionalmente viável, para a elaboração das escalas de trabalho de pessoal de enfermagem, capaz de gerar elementos para avaliação de sua influência no desempenho operacional do hospital, torna-se importante. A contribuição desta pesquisa é apresentar um modelo matemático, baseado em restrições que permitam a elaboração de escalas para equipes de enfermagem com suporte computacional. Além disso, o estudo apresenta elementos para avaliação do Quadro de Lotação da equipe de trabalho. A tarefa de escalar pessoal é responsável por consumir muito tempo e nem sempre garantir o cumprimento das legislações e normas vigentes, visto ser normalmente um processo manual, sem uso de ferramentas computacionais personalizáveis às particularidades de cada hospital. O estudo foi realizado a partir de dados levantados junto a Unidade de Internação de um hospital em Porto Alegre/RS. Foram obtidas, no estudo de campo, informações relativas às necessidades da escala de trabalho nesta realidade particular, além do levantamento de indicadores de desempenho. Foram realizados experimentos para avaliar a influência das variáveis de decisão e parâmetros do Algoritmo empregado para a obtenção da solução. Na validação do modelo foram utilizados casos de teste hipotéticos, baseados em dados reais levantados no estudo. Com a aplicação do modelo, as escalas de trabalho puderam ser obtidas de forma a atender as necessidades dos setores hospitalares proporcionando menor impacto no desempenho operacional do hospital. / The operating performance of hospitals have directly related of rostering colaboration. Thus, the construction of the same task requires attention to the many details to obtain good results. In this context, a model computationally feasible to prepare the rostering nursing, capable of generating elements to assess its influence on the hospital's operating performance, it is important. The contribution of this research is to present a formal mathematical model, based on restrictions to allow the construction of scheduling nursing with computer support. In addition, the study presents elements for evaluating the Board's work team Capacity. Personal climbing task is responsible for time-consuming and not always ensure compliance with existing laws and regulations, as it is usually a manual process, without the use of customizable computational tools to the particularities of each hospital. The study was conducted from data collected at the inpatient unit of a hospital in Porto Alegre/ RS. They were obtained from the information field study on working range of needs in this particular reality, beyond the performance indicators survey. Experiments were conducted to evaluate the effect of the decision variables and the influence of the parameters of the Evolutionary Algorithm in the final solution. To validate the model we used cases of hypothetical test, based on real data collected in the study. With the application of the model, working scales can be obtained to meet the needs of hospital departments providing less negative impact on the hospital's operating performancer.

Made available in DSpace on 2014-12-17T14:53:08Z (GMT). No. of bitstreams: 1 MiriamKR_DISSERT.pdf: 394744 bytes, checksum: 86723bb7ac2b34e7cfb757711ca68d77 (MD5) Previous issue date: 2012-06-25 / Coordena??o de Aperfei?oamento de Pessoal de N?vel Superior / This work aims to "build" rostering urban bus crews to minimize the cost of overtime. For this purpose a mathematical model was developed based on case study in an urban transport company in the metropolitan region of Natal. This problem is usually known in the literature as the Crew Scheduling Problem (CSP) and classified as NP-hard. The mathematical programming takes into account constraints such as: completion of all trips, daily and maximum allowable range of home and / or food. We used the Xpress-MP software to implement and validate the proposed model. For the tested instances the application of the model allowed a reduction in overtime from 38% to 84% / Este trabalho tem por objetivo construir escalas de servi?os de tripula??es de ?nibus urbano de forma a minimizar o custo com horas extras. Para tanto desenvolveu-se um modelo matem?tico por meio de um estudo de caso em uma empresa de transporte urbano na regi?o metropolitana de Natal. Este problema, de uma maneira geral, ? conhecido na literatura como Problema de Programa??o de Tripula??o (PPT) e classificado como NP-dif?cil. A programa??o matem?tica contempla restri??es tais como: realiza??o de todas as viagens, jornada di?ria m?xima permitida e intervalo de repouso e/ou alimenta??o. Foi utilizado o aplicativo Xpress-MP para implementar e validar o modelo proposto. Para as inst?ncias testadas o modelo apresentou uma redu??o da hora extra entre 38% e 84%.

The logistics industry is one of the most important industries, being an important component of a company's activities. In recent years, the industry has gone through a large number of changes. These had the purpose of making the processes more efficient in order to reduce costs or to increase the degree of competitiveness. A very important process in the logistics industry is the allocation of roles in warehouses. Not using the available resources to their maximum capacity leads to an increase in the processing time of the units as well as to a decrease in the profitability of the companies. Allocation is done manually in the warehouse, being a complex task in which many constraints and variations must be taken into account. AI technologies and IoT systems have the potential to simplify the process of assigning workers. For this research, a warehouse in Germany operated by the Amazon company was studied. In addition to the benefits that this transition can have from a social point of view, helping to equalize the roles of high complexity and the appropriate rotation of jobs to ensure equal opportunities for development, there is an opportunity to reduce the costs associated with this process by up to 90%. This last aspect is dependent on the degree of modernization of the warehouse and on the way in which the role allocation process is carried out at the time. Most of the IoT systems necessary for the digitization and automation of this process are already implemented in warehouses. However, a common architecture is needed to be able to guarantee the integration and compatibility of the systems. The main blocker in this transition remains public opinion and financial factors, companies with a large volume of units potentially benefiting from economies of scale. / Logistikbranschen är en av de viktigaste branscherna och är en viktig del av ett företags verksamhet. De senaste åren har branschen genomgått ett stort antal förändringar. Dessa hade till syfte att effektivisera processerna för att minska kostnaderna eller öka graden av konkurrenskraft. En mycket viktig process inom logistikbranschen är fördelningen av roller i lager. Att inte använda de tillgängliga resurserna till sin maximala kapacitet leder till en ökad handläggningstid för enheterna samt till en minskad lönsamhet för företagen. Allokering sker manuellt i lagret, vilket är en komplex uppgift där många begränsningar och variationer måste beaktas. AI-teknik och IoT-system har potential att förenkla processen för att tilldela arbetare. För denna forskning studerades ett lager i Tyskland som drivs av Amazon-företaget. Förutom de fördelar som denna övergång kan ha ur en social synvinkel, hjälpa till att utjämna rollerna med hög komplexitet och lämplig rotation av jobb för att säkerställa lika möjligheter till utveckling, finns det en möjlighet att minska kostnaderna förknippade med denna process med upp till 90 %. Den sista aspekten är beroende av graden av modernisering av lagret och på det sätt på vilket rollfördelning processen går till vid tillfället. De flesta av de IoT-system som krävs för digitalisering och automatisering av denna process är redan implementerade i lager. Det behövs dock en gemensam arkitektur för att kunna garantera systemets integration och kompatibilitet. Den huvudsakliga blockeraren i denna övergång är fortfarande den allmänna opinionen och finansiella faktorer, företag med en stor volym av enheter som potentiellt kan dra nytta av stordriftsfördelar.