Denna studie undersöker vilken inverkan voice picking-systemet har på effektiviteten i orderplockningsprocessen samt hur arbetsmiljön påverkas med Baxter som fallföretag. Under det senaste decenniet har ny informationsteknologi haft stor inverkan på logistikbranschen. En av dessa nya teknologier är voice picking-systemet. I detta arbete behandlas faktorer såsom effektivitet och produktivitet, flaskhalsar, mänskliga fel, uttagningsprinciper, lagerstruktur och rutter, arbetsmiljö samt för- och nackdelar som verksamheter bör ta hänsyn till vid användning av ett voice picking-system. Det visade att systemet är effektivt men även att det finns nackdelar att beakta. Metoden som används är en kvalitativ fallstudie. Under arbetets gång genomfördes en observation och åtta intervjuer med anställda som hade olika arbetsroller på Baxters lagerverksamhet i Lund. Samtlig data analyserades därefter med hjälp av studiens teoretiska ramverk och empiri för att kunna avgöra vilken påverkan voice picking-systemet har. Studiens resultat visar att ett voice picking-system kan bidra till en effektivare orderplockningsprocess. Dock kan det också leda till minskad arbetsmotivation på grund av bristande möjlighet till beslutsfattande och ansvarstagande för medarbetarna. / This study investigates what impact the voice picking-system has on the efficiency of the order picking process and how the work environment is affected with Baxter as a case company. Over the past decade, new information technology has had a major impact on the logistics industry. One of these new technologies is the voice picking-system. In this study will factors such as efficiency and productivity, bottlenecks, human error, picking principles, warehouse structure and routes, working environment and pros and cons that businesses should consider when using a voice picking system be treated. It turned out that the voice picking-system is effective, but that there are also disadvantages to consider. The method used in this survey is a qualitative case study. During the study, one observation and eight interviews were accomplished with employees who had different roles at Baxter's warehouse in Lund. All data was then analyzed using the study's theoretical framework and the empirical evidence to determine the impact of the voice picking system. The study's results show that a voice picking system can contribute to a more efficient order picking process. However, it can also lead to reduced work motivation due to a lack of opportunities for decision-making and responsibility for the employees.

Esta tese integra dois problemas de áreas distintas e interdependentes intitulados de Sequenciamento Otimizado de Faturamento (SOF) e Sequenciamento Otimizado de Coleta (SOC). Abordados de forma disjunta pelos pesquisadores, o SOF refere-se a um problema de maximização do faturamento e o SOC consiste de uma variação do Order Batching and Sequencing Problem (OBSP). Fundamentados por pressupostos práticos e científicos, o SOF/SOC retratam o cotidiano dos processos de faturamento e picking de um típico Armazém de Distribuição (AM). No SOF a demanda é estocástica e os faturamentos ocorrem a partir de janelas de tempo variáveis ajustadas para evitar o tardiness mediante a priorização das datas de atendimento pela regra Earliest Due Date (EDD). No SOC o picking é manual e enquadra-se na categoria picker-to-parts (low level) com pick-and-sort utilizando um trolley que é transportado pelo operador ao longo das ruas do AM. Neste contexto, esta tese tem como objetivo desenvolver uma ferramenta de gestão que integre e apresente soluções otimizadas para o SOF/SOC. A perspectiva de integração do SOF/SOC dar-se-á mediante à formulação de dois Algoritmos Genéticos (AGs) nomeados de AG-SOF e AG-SOC. Assim, o enfoque desta pesquisa está na avaliação da eficácia prática do AG-SOF/AG-SOC em resolver problemas reais do SOF/SOC. A eficácia do AG-SOF é comparada à um Algoritmo Guloso Iterativo (AI-SOF) enquanto que a predileção pelo AG-SOC é justificada pela natureza NP-hard do SOC. As experimentações para problemas de diferentes níveis de complexidade demonstraram que os algoritmos satisfazem todas as regras, restrições e variáveis decisórias obtendo soluções de qualidade satisfatória para qualquer categoria do SOF/SOC. O AISOF/ AG-SOF lidam com as restrições de estoque e as possibilidades de faturar pedidos parciais para maximizar o Faturamento Total (FT). Apesar de obterem soluções com a mesma qualidade, o AI-SOF tem desempenho superior ao AG-SOF que é, em termos de Tempo de Processamento Computacional (TPC), limitado às categorias de médio porte do SOF. O AG-SOC é composto pela iteração de dois AGs (AGLOTE e AGPCV) que minimizam o Custo Total das Operações de Picking (CT). Logo, o AGLOTE agrupa os SKUs (Stock Keeping Units) dos diferentes pedidos em múltiplos lotes pela restrição de carga dos trolleys de forma a reduzir o Número de Viagens de Coleta (NVC) e define a sequência de coleta por meio de lotes prioritários para evitar o Atraso no Atendimento dos Pedidos (AAP). O AGPCV faz a roteirização dos lotes dentro do AM de modo que impeça a ocorrência de avarias aos SKUs frágeis e minimize a Distância Total das Rotas (DTR) e o Tempo Total de Picking (TTP). Evidenciou-se que para problemas de complexidade superior os lotes são mais homogêneos, nos quais o Desvio Padrão é pequeno e o Coeficiente de Variação é de 11,22% a 25,20% para a DTR. Para ambientes reais em que se utiliza janelas de tempo e logs de processamentos para lotes off-lines) a combinação do AI-SOF/AGSOC provê soluções otimizadas em tempo e qualidade satisfatória ao SOF/SOC. Em suma, esta pesquisa foi além das abordagens existentes para preencher um gap na literatura e prover uma importante contribuição à prática da otimização do SOF/SOC. É possível conclui que a integração do AI-SOF/AGSOC é capaz de maximizar o faturamento e melhorar a produtividade de forma a minimizar os tempos e custos operacionais de picking do AM. / This thesis integrates two problems from distinct and independent areas called Optimal Sequencing Billing (OSB) and Optimal Picking Sequencing (OPS). Studied separately by researchers, OBS refers to a billing maximization problem and OPS is a variation of the Order Batching and Sequencing Problem (OBSP). Based on practical and scientific assumptions, OSB/OPS portray the picking daily routine in a typical Distribution Warehouse (WA). In OSB, the demand is stochastic and billings occur based on variables time windows that are adjusted to avoid tardiness by prioritizing the service dates based on the Earliest Due Date (EDD) rule. In OPS, picking is manual and falls into the low-level picker-to-parts category, and it uses a trolley that is pushed by an employee along WA aisles. In this context, this thesis has the objective of developing a management tool that can integrate and provide optimal solutions for OSB/OPS. The perspective of integrating OSB/OPS can be achieved through the formulation of two Genetic Algorithms (GAs) called GA-OSB and GA-OPS. Therefore, the focus of this research is to assess GA-OSB/GA-OPS practical efficiency to solve actual OSB/OPS problems. GA-OSB efficiency is compared to an Iterative Greedy Algorithm (IA-OSB) whereas the preference for GA-OPS is justified by the NP-hard nature of OPS. Experiments for problems of different complexity levels showed that algorithms satisfy every rule, restriction and decision variable and provide satisfactory solutions for any OSB/OPS category. IA-OSB/GA-OSB deal with inventory restrictions and the possibility of billing partial orders to maximize Total Billing (TB). Although they also provide quality solutions, IA-OSB performance is better than GA-OSB performance which is limited to OSB medium-sized categories in terms of Computational Processing Time (CPT). GA-OPS comprises the iteration of two GAs (GABATCH and GATSP) that minimize the Total Cost of Picking Operations (TC). Therefore, GABATCH groups SKUs (Stock Keeping Units) of different orders into multiple lots according to trolley load restrictions so as to reduce the Number of Picking Travels (NPT). It also defines a picking sequence by means of priority lots to avoid Tardiness in Customer Orders (TCO). GATSP maps out the routes of lots inside the WA in order to prevent damages to fragile SKUs and to minimize Total Route Distance (TRD) as well as Total Picking Time (TPT). It was evidenced that, for problems of higher complexity, lots are more homogeneous where the Standard Deviation is small and the Coefficient of Variation (CV) ranges from 11.22% to 25.20% to the TRD. For actual environments where time windows and processing logs are used for off-line lots, the IA-OSB/GA-OPS integration provides optimal time solutions and satisfactory quality to OSB/OPS. In short, this research has gone beyond existing approaches to fill a gap in the literature and provide an important contribution to the practice of optimal OSB/OPS. It can be concluded that the integration of IA-OSB/GA-OPS can maximize billing and improve productivity in order to minimize picking operational time and costs in a WA.

Esta tese integra dois problemas de áreas distintas e interdependentes intitulados de Sequenciamento Otimizado de Faturamento (SOF) e Sequenciamento Otimizado de Coleta (SOC). Abordados de forma disjunta pelos pesquisadores, o SOF refere-se a um problema de maximização do faturamento e o SOC consiste de uma variação do Order Batching and Sequencing Problem (OBSP). Fundamentados por pressupostos práticos e científicos, o SOF/SOC retratam o cotidiano dos processos de faturamento e picking de um típico Armazém de Distribuição (AM). No SOF a demanda é estocástica e os faturamentos ocorrem a partir de janelas de tempo variáveis ajustadas para evitar o tardiness mediante a priorização das datas de atendimento pela regra Earliest Due Date (EDD). No SOC o picking é manual e enquadra-se na categoria picker-to-parts (low level) com pick-and-sort utilizando um trolley que é transportado pelo operador ao longo das ruas do AM. Neste contexto, esta tese tem como objetivo desenvolver uma ferramenta de gestão que integre e apresente soluções otimizadas para o SOF/SOC. A perspectiva de integração do SOF/SOC dar-se-á mediante à formulação de dois Algoritmos Genéticos (AGs) nomeados de AG-SOF e AG-SOC. Assim, o enfoque desta pesquisa está na avaliação da eficácia prática do AG-SOF/AG-SOC em resolver problemas reais do SOF/SOC. A eficácia do AG-SOF é comparada à um Algoritmo Guloso Iterativo (AI-SOF) enquanto que a predileção pelo AG-SOC é justificada pela natureza NP-hard do SOC. As experimentações para problemas de diferentes níveis de complexidade demonstraram que os algoritmos satisfazem todas as regras, restrições e variáveis decisórias obtendo soluções de qualidade satisfatória para qualquer categoria do SOF/SOC. O AISOF/ AG-SOF lidam com as restrições de estoque e as possibilidades de faturar pedidos parciais para maximizar o Faturamento Total (FT). Apesar de obterem soluções com a mesma qualidade, o AI-SOF tem desempenho superior ao AG-SOF que é, em termos de Tempo de Processamento Computacional (TPC), limitado às categorias de médio porte do SOF. O AG-SOC é composto pela iteração de dois AGs (AGLOTE e AGPCV) que minimizam o Custo Total das Operações de Picking (CT). Logo, o AGLOTE agrupa os SKUs (Stock Keeping Units) dos diferentes pedidos em múltiplos lotes pela restrição de carga dos trolleys de forma a reduzir o Número de Viagens de Coleta (NVC) e define a sequência de coleta por meio de lotes prioritários para evitar o Atraso no Atendimento dos Pedidos (AAP). O AGPCV faz a roteirização dos lotes dentro do AM de modo que impeça a ocorrência de avarias aos SKUs frágeis e minimize a Distância Total das Rotas (DTR) e o Tempo Total de Picking (TTP). Evidenciou-se que para problemas de complexidade superior os lotes são mais homogêneos, nos quais o Desvio Padrão é pequeno e o Coeficiente de Variação é de 11,22% a 25,20% para a DTR. Para ambientes reais em que se utiliza janelas de tempo e logs de processamentos para lotes off-lines) a combinação do AI-SOF/AGSOC provê soluções otimizadas em tempo e qualidade satisfatória ao SOF/SOC. Em suma, esta pesquisa foi além das abordagens existentes para preencher um gap na literatura e prover uma importante contribuição à prática da otimização do SOF/SOC. É possível conclui que a integração do AI-SOF/AGSOC é capaz de maximizar o faturamento e melhorar a produtividade de forma a minimizar os tempos e custos operacionais de picking do AM. / This thesis integrates two problems from distinct and independent areas called Optimal Sequencing Billing (OSB) and Optimal Picking Sequencing (OPS). Studied separately by researchers, OBS refers to a billing maximization problem and OPS is a variation of the Order Batching and Sequencing Problem (OBSP). Based on practical and scientific assumptions, OSB/OPS portray the picking daily routine in a typical Distribution Warehouse (WA). In OSB, the demand is stochastic and billings occur based on variables time windows that are adjusted to avoid tardiness by prioritizing the service dates based on the Earliest Due Date (EDD) rule. In OPS, picking is manual and falls into the low-level picker-to-parts category, and it uses a trolley that is pushed by an employee along WA aisles. In this context, this thesis has the objective of developing a management tool that can integrate and provide optimal solutions for OSB/OPS. The perspective of integrating OSB/OPS can be achieved through the formulation of two Genetic Algorithms (GAs) called GA-OSB and GA-OPS. Therefore, the focus of this research is to assess GA-OSB/GA-OPS practical efficiency to solve actual OSB/OPS problems. GA-OSB efficiency is compared to an Iterative Greedy Algorithm (IA-OSB) whereas the preference for GA-OPS is justified by the NP-hard nature of OPS. Experiments for problems of different complexity levels showed that algorithms satisfy every rule, restriction and decision variable and provide satisfactory solutions for any OSB/OPS category. IA-OSB/GA-OSB deal with inventory restrictions and the possibility of billing partial orders to maximize Total Billing (TB). Although they also provide quality solutions, IA-OSB performance is better than GA-OSB performance which is limited to OSB medium-sized categories in terms of Computational Processing Time (CPT). GA-OPS comprises the iteration of two GAs (GABATCH and GATSP) that minimize the Total Cost of Picking Operations (TC). Therefore, GABATCH groups SKUs (Stock Keeping Units) of different orders into multiple lots according to trolley load restrictions so as to reduce the Number of Picking Travels (NPT). It also defines a picking sequence by means of priority lots to avoid Tardiness in Customer Orders (TCO). GATSP maps out the routes of lots inside the WA in order to prevent damages to fragile SKUs and to minimize Total Route Distance (TRD) as well as Total Picking Time (TPT). It was evidenced that, for problems of higher complexity, lots are more homogeneous where the Standard Deviation is small and the Coefficient of Variation (CV) ranges from 11.22% to 25.20% to the TRD. For actual environments where time windows and processing logs are used for off-line lots, the IA-OSB/GA-OPS integration provides optimal time solutions and satisfactory quality to OSB/OPS. In short, this research has gone beyond existing approaches to fill a gap in the literature and provide an important contribution to the practice of optimal OSB/OPS. It can be concluded that the integration of IA-OSB/GA-OPS can maximize billing and improve productivity in order to minimize picking operational time and costs in a WA.

No description available.

This is a bachelor's thesis in the area of electrical engineering as well as automation. This thesis investigates motor choice, sensor choices, and the methodology of automating a grocery store picking system prototype. The general structure of the system is composed out of two shelfs with a plane, between them, that is movable in the vertical direction. On the movable plane a conveyor belt was installed. We evaluated the motor choice for the application and the safety aspects of the system to gather a general understanding of potential risk factors involved in realizing a system that can work autonomously. It includes both electrical safety factors as well as mechanical safety factors. Additionally, we considered safety precautions to prevent malfunctions that can lead to these risks. To realize the prototype, a prestudy to investigate motor alternatives was conducted to determine the best motor choice for the application. The chosen motor was a DC steppermotor with snail gearbox. This will ensure simple regulation of the motor to sufficiently handle the weights the system needs to manage. The snail gearbox can eliminate some mechanical risk factors due to its self-locking mechanism, ensuring that if there is an electrical malfunction, such as a power failure, the system will not collapse on itself and drop the payload flat on the floor. Also, a model of the large prototype was made in a 3D-drawing software, which was then 3D-printed and set up to mimic the large prototype. The goal of the miniature model is to test the functionality and evaluate if the algorithms described in this thesis can be preserved and re-used for the large system and prototype.