Two important areas of decision-making in distribution system design involve facility location and inventory policy determination. Facility location analyzes questions such as how many facilities should be opened, where they should be located, and which customers should be assigned to which DCs. Inventory policy determination involves more tactical decisions such as the order quantities and frequencies at each level or echelon in the network. It is believed that these two decisions can influence each other significantly. Including a multi-echelon inventory policy decision in a location analysis allows a user to capitalize on the strengths that each DC has to offer (e.g., lower labor rates, land costs, etc.). Likewise, when the locations of two facilities are known, a multi-echelon inventory policy can be designed better to incorporate the exact lead times and fixed costs between the facilities at each level of the system. Despite this, the two problems are typically solved independently. This research addresses these problems together and investigates different heuristic methods for solving a combined inventory-location model. We begin by presenting the background and formulation for each problem. These formulations are then combined to show how the two problems can be mathematically formulated together. Rather than solve the problem exactly, two heuristic methods using different philosophies are tested. We apply these heuristic methods to the combined inventory-location problem to determine how much we can improve distribution network design solutions and what type of heuristic methodology is most effective in gaining these improvements. Our results show that the combined inventory-location model is capable of improving on the solutions obtained by a location model with a fixed inventory policy. The improvement based on the data sets tested in this research was approximately $60,000. However, in cases where the inventory costs are a larger portion of the total cost, the improvement made by the inventory-location model increased to over $1,000,000. We also found that our second heuristic method tested provided statistically significant improved results over our first heuristic method. Moreover, the second heuristic method typically ran 67% faster. The improved results, although small in a relative sense (the average improvement was 0.18%), would still represent a large absolute improvement in supply chain costs. As much as $174,000 was saved in the data sets tested for this research. / Master of Science

The planning of an Aftermarket Supply Chain is a very complex task. This is due to an unpredictable demand which is driven by the need for maintenance and repair. This drive translates to a high variety of lead times, a large number of stock-keeping units (SKUs) and the capacity to deliver spare parts during its full lifecycle. With all these complexities in place, optimizing and parametrizing the planning process is a difficult and time-consuming task. Moreover, the current optimization tool focuses only on one node (each warehouse individually) of the whole Supply Chain, without considering the information such as inventory levels of the other nodes. Hence, the Supply Chain is not completely connected, making it difficult to get a better understanding of the system performance to identify cost draining areas. This leads to capital being tied up in the upper stream of the Supply Chain and later adding unnecessary costs like high inventory costs, rush freight costs, return or scrapping cost. In this study, Discrete Event Simulation (DES) is explored as an additional optimization tool that could analyse and improve the performance of the whole Supply Chain. To do that, the functioning of a node is modelled by replicating the logics behind the flow of material, which includes analysing some manual workflows which are currently present. In Addition, all the information needed from the orders, order lines and parts are mapped. The later part of the study aims to connect all the nodes to form a whole overview of the Supply Chain and further perform optimizations globally.  As an outcome, Multi-Echelon Inventory Optimization has been performed on the whole Supply Chain after connecting all the nodes and thus getting an overview. Furthermore, the impact of different parameters has been studied on the whole model to understand the sensitivity of parameters such as variations in lead time and demand. Finally, different what-if scenarios such as COVID and problems with delay in suppliers were studied, which could help understand the impact of unforeseen situations. / Planeringen av en eftermarknadskedja är en mycket komplex uppgift. Detta beror på en oförutsägbar efterfrågan som drivs av behovet av underhåll och reparation. Enheten översätter till många olika ledtider, ett stort antal lagerhållningsenheter (SKU) och kapacitet att leverera reservdelar under hela dess livscykel. Med alla dessa komplexiteter på plats är optimering och parametrering av planeringsprocessen en svår och tidskrävande uppgift. Dessutom fokuserar det nuvarande optimeringsverktyget bara på en nod (varje lager separat) i hela leveranskedjan utan att beakta informationen som lagernivåerna för de andra noderna. Därför är försörjningskedjan inte helt ansluten, vilket gör det svårt att få en bättre förståelse för systemets prestanda för att identifiera kostnadsavtappningsområden. Detta leder till att kapital binds i den övre strömmen i försörjningskedjan och senare lägger till onödiga kostnader som höga lagerkostnader, snabba fraktkostnader, retur- eller skrotningskostnader. I denna studie undersöks Discrete Event Simulation (DES) som ett ytterligare optimeringsverktyg som kan analysera och förbättra prestanda för hela försörjningskedjan. För att göra det modelleras en nods funktion genom att replikera logiken bakom materialflödet, vilket inkluderar analys av några manuella arbetsflöden som för närvarande finns. Dessutom kartläggs all information som behövs från beställningar, orderrader och delar. Den senare delen av studien syftar till att ansluta alla noder för att bilda en hel översikt över försörjningskedjan och ytterligare utföra optimeringar globalt. Som ett resultat har Multi-Echelon Lageroptimering utförts i hela försörjningskedjan efter att alla noder har anslutits och därmed fått en översikt. Dessutom har effekterna av olika parametrar studerats på hela modellen för att förstå känsligheten hos parametrar som variationer i ledtid och efterfrågan. Slutligen studerades olika tänkbara scenarier som COVID och problem med förseningar hos leverantörer, vilket kan hjälpa till att förstå effekterna av oförutsedda situationer.

O foco dessa dissertação está na gestão de estoques em sistemas multicamadas. O modelo de simulação construído considera o desempenho histórico de uma cadeia de distribuição de medicamentos versus a política de gestão proposta. Os objetivos principais do trabalho são: a avaliação quantitativa da política de cálculo de necessidades aplicada na gestão de sistemas multicamadas, a verificação da importância da cooperação entre os elos da cadeia de distribuição para gestão do fluxo de materiais e o estudo das curvas que representam o dilema ‘nível de serviço’ versus ‘custo total da cadeia’. O modelo considera o fluxo de materiais a partir do estoque em processo do laboratório até a venda para as farmácias, que pode ser aproximada pela demanda dos medicamentos visto que os estoques das farmácias são relativamente pequenos e constantes ao longo do tempo. A modelagem de custos leva em conta o custo de estoque do laboratório e dos distribuidores, o custo de pedido dos distribuidores, o custo de transporte, o custo de entrega com atraso do laboratório e o custo de venda perdida dos distribuidores. Especificamente, são considerados cenários com e sem o compartilhamento de informações entre as empresas da cadeia. Consideram-se também cenários com e sem sazonalidade na demanda. Inicialmente, cada cenário simulado é preparado com a definição do ‘período transitório da simulação’, ‘horizonte de simulação’ e ‘número de réplicas’ necessárias. Depois disso, é feito o delineamento de experimentos para identificar quais variáveis de decisão têm efeito significativo sobre o custo total da cadeia. Finalmente, é feita uma busca da parametrização de cada um dos cenários que apresente o melhor custo total da cadeia. Os resultados da simulação mostraram que as práticas atualmente empregadas na gestão dos estoques das empresas podem ter seu desempenho melhorado com a utilização da política simulada na pesquisa. Os cenários nos quais se considerou o compartilhamento de informações tiveram desempenhos semelhantes aos sem esse compartilhamento. Dessa forma, para a política simulada, conclui-se que o valor do compartilhamento de informações foi relativamente pequeno. Nas simulações pode-se verificar a melhoria simultânea do nível de serviço e do nível de estoques da cadeia. Isso mostra que a política simulada mudou o dilema (trade-off) que interliga antagonicamente essas duas características de desempenho do sistema. / This dissertation focuses on the inventory management for multi-echelon systems. The simulation model proposes a new inventory management policy and compares it to the historical performance of a medicine supply chain. The main objectives of this research are: the quantitative analysis of the method used for the calculation of material requirements in multi-echelon systems; the verification of the importance of the cooperation between the components of the supply chain to the material flow management; and the analysis of the curves which represent the trade-off between ‘service level’ and ‘total cost’. The model takes into account the material flow from the laboratory’s ‘work in process’ up to the sales to the drugstores, which can be approximated by the demand of the final consumer, since the inventory kept by drugstores is relatively small and constant in time. The costs are calculated considering: the laboratory’s and distributors’ inventory costs, the cost of orders from the distributors, the transportation cost, the laboratory’s cost of late delivery and the distributors’ cost of lost sales. Scenarios with and without the sharing of information between the components of the supply chain were both considered. The same is true for scenarios with and without seasonality in the demand. Initially, each scenario was prepared with the definition of the ‘warm-up’ period, the simulation horizon and the amount of required replications. Secondly, the design of experiments (DOE) was done in order to determine which decision variables have influence on the supply chain total cost. Finally, each scenario was tested with many different parameters in order to find the lowest cost for the supply chain. The simulation results have showed that the procedures currently applied for the inventory management can have their performance improved by the use of the policy proposed in this research. The results for the scenarios with the sharing of information were similar to the ones for the scenarios without the sharing. For that reason, we can conclude that, for the proposed inventory management policy, the value of the sharing of information through the supply chain was relatively small. In the simulations, both the ‘service level’ and the ‘total cost’ have improved. In this manner, it can be said that the new policy has improved this trade-off.

O foco dessa dissertação está na gestão de estoques em sistemas multicamadas. O modelo de simulação construído considera o desempenho histórico de uma cadeia de distribuição de medicamentos versus a política de gestão proposta. Os objetivos principais do trabalho são: a avaliação quantitativa da política de cálculo de necessidades aplicada na gestão de sistemas multicamadas, a verificação da importância da cooperação entre os elos da cadeia de distribuição para gestão do fluxo de materiais e o estudo das curvas que representam o dilema ‘nível de serviço’ versus ‘custo total da cadeia’. O modelo considera o fluxo de materiais a partir do estoque em processo do laboratório até a venda para as farmácias, que pode ser aproximada pela demanda dos medicamentos visto que os estoques das farmácias são relativamente pequenos e constantes ao longo do tempo. A modelagem de custos leva em conta o custo de estoque do laboratório e dos distribuidores, o custo de pedido dos distribuidores, o custo de transporte, o custo de entrega com atraso do laboratório e o custo de venda perdida dos distribuidores. Especificamente, são considerados cenários com e sem o compartilhamento de informações entre as empresas da cadeia. Consideram-se também cenários com e sem sazonalidade na demanda. Inicialmente, cada cenário simulado é preparado com a definição do ‘período transitório da simulação’, ‘horizonte de simulação’ e ‘número de réplicas’ necessárias. Depois disso, é feito o delineamento de experimentos para identificar quais variáveis de decisão têm efeito significativo sobre o custo total da cadeia. Finalmente, é feita uma busca da parametrização de cada um dos cenários que apresente o melhor custo total da cadeia. Os resultados da simulação mostraram que as práticas atualmente empregadas na gestão dos estoques das empresas podem ter seu desempenho melhorado com a utilização da política simulada na pesquisa. Os cenários nos quais se considerou o compartilhamento de informações tiveram desempenhos semelhantes aos sem esse compartilhamento. Dessa forma, para a política simulada, conclui-se que o valor do compartilhamento de informações foi relativamente pequeno. Nas simulações pode-se verificar a melhoria simultânea do nível de serviço e do nível de estoques da cadeia. Isso mostra que a política simulada mudou o dilema (trade-off) que interliga antagonicamente essas duas características de desempenho do sistema. / This dissertation focuses on the inventory management for multi-echelon systems. The simulation model proposes a new inventory management policy and compares it to the historical performance of a medicine supply chain. The main objectives of this research are: the quantitative analysis of the method used for the calculation of material requirements in multi-echelon systems; the verification of the importance of the cooperation between the components of the supply chain to the material flow management; and the analysis of the curves which represent the trade-off between ‘service level’ and ‘total cost’. The model takes into account the material flow from the laboratory’s ‘work in process’ up to the sales to the drugstores, which can be approximated by the demand of the final consumer, since the inventory kept by drugstores is relatively small and constant in time. The costs are calculated considering: the laboratory’s and distributors’ inventory costs, the cost of orders from the distributors, the transportation cost, the laboratory’s cost of late delivery and the distributors’ cost of lost sales. Scenarios with and without the sharing of information between the components of the supply chain were both considered. The same is true for scenarios with and without seasonality in the demand. Initially, each scenario was prepared with the definition of the ‘warm-up’ period, the simulation horizon and the amount of required replications. Secondly, the design of experiments (DOE) was done in order to determine which decision variables have influence on the supply chain total cost. Finally, each scenario was tested with many different parameters in order to find the lowest cost for the supply chain. The simulation results have showed that the procedures currently applied for the inventory management can have their performance improved by the use of the policy proposed in this research. The results for the scenarios with the sharing of information were similar to the ones for the scenarios without the sharing. For that reason, we can conclude that, for the proposed inventory management policy, the value of the sharing of information through the supply chain was relatively small. In the simulations, both the ‘service level’ and the ‘total cost’ have improved. In this manner, it can be said that the new policy has improved this trade-off.

The supply function has an important role to support the business to create a customer value. Two important parts of this process is to have the warehouses and production sites in the right location and to have the right items stocked at the right level.   This thesis is concerned with those two parts of the supply chain management. Three different areas of inventory control are dealt with. In the first part we consider the classical dynamic lot size problem without backlogging. The second part deals with estimation of holding and shortage costs in two-level distribution inventory systems. In the third part of the thesis we consider the localisation problem in a multi-level supply network system where items are consolidated at a warehouse and distributed to customers on routes.   Within the area of inventory control we have evaluated a method earlier suggested by Axsäter (1988), the method is evaluated using a set of test problems and compared other heuristic methods, including the well-known Silver-Meal’s method (Silver and Meal, 1973).  The result shows that the method suggested by Axsäter does perform better than the other methods. In the latest contribution we point to the important differences between Least Period Cost and Silver-Meal when several periods have zero demand. In the area of inventory control we have also studied a two-echelon inventory system where we present methods for estimating the shortage- and stockholding costs in such inventory systems.   The second part subject of the thesis concerns supply network optimization. We present a MIP formulation of the problem and evaluate in detail the approximation of the distribution cost when customers are delivered on multi-stop routes. An improved method for estimating the distribution is presented.   Besides this introductory overview five research papers are included in the thesis. The first and the last paper consider evaluation of dynamic lot sizing heuristics. The second and third paper deals with cost evaluation of a stochastic two-echelon inventory system and the forth paper with evaluation of methods for estimating distribution costs in a supply network.

No description available.