Criação do Cenário

Criação do Cenário

1. Simulator Domain

   • Seleção dos campos da base de cenário utilizada
   • SimulateItem
     • Analyze
       • Consumo
       • Lead Time
     • Processo
       • Fonte de informação de consumo
       • Fonte de informação de Lead Time
       • Organiza Nível de serviço
       • Checa o tipo de simulação a usar
       • Execução dos métodos de Simulação
         • Caso simulationModel = SimulationTypeModel.Poisson
           • -Inputs: lifeCycle, lots, leadTimes, leadTimeAverage, leadTimeDeviation, minimumBuyLot, multipleBuyLot, consumptionLot, orderCost, stockCost, unitaryPrice.
           • Output: Warning, Minimum, Maximum, BuyLot (lotsRestriction), LotCoverage(Int(lotsRestriction / demandAverage)), P1, P2, SafetyStock (securityStock), AvgStock,AvgValueInStock (averageStock * unitaryPrice) e AmountOfResupply
           • Validar esse uso do ConsumptionLot. Ao meu ver não faz sentido
           • Define as variáveis
             • Todas as variáveis criadas: -lots = itemsinOperation/ConsumptionLot-demandAverage = lots / lifeCycle-annualDemand = demandAverage * 12-lotsRestriction = Lot(annualDemand, orderCost, stockCost, minimumBuyLot, multipleBuyLot, minimumCoverage, maximumCoverage, demandAverage);-leadTimeAvg = leadTimeAverage-leadTimeStandardDeviation = leadTimeDeviation-leadTimeDemandAverage = leadTimeAvg * demandAverage
           • Cálculo do ponto do pedido
             • Iteração entre o requestPoint e os níveis de serviço : expectedServiceLevel1 e expectedServiceLevel2
             • Explicação: Como na modelagem de poisson não há a padronização do modelo estatístico não é possível definir k usado na fórmula s = udL + k*stdevL, por esse motivo, é feito a iteração para definir o request point que atende o nível de serviço desejado.
           • Resultados
           • Validações e variações finais
         • Caso simulationModel = SimulationTypeModel.Bernoulli
         • Caso simulationModel = SimulationTypeModel.GeneralizedCompound
         • Caso simulationModel = SimulationTypeModel.GeneralizedCompound2
           • consumptions[],minimalLimit, averageConsumption,consumptionDeviation, consumptionRate, consumptionType, leadTimes[][], leadTimeAverage, leadTimeDeviation, minimumBuyLot, multipleBuyLot, consumptionLot, orderCost, stockCost,unitaryPrice, -wantedP1,wantedP2, minimumCoverage, maximumCoverage = 0;
           • Define as variáveis
             • leadTimeLength = leadTimes[1].Length
               • Qntd eventos Lead time
             • Variáveis zeradas(0)
               • consumptionLength, nonConsumptionProbability, demandAverage, calculatedDemandAvg, demandStandardDeviation, leadTimeNonConsumptionProb
             • Cálculo Média/Desvpad
               • Cálculo da média e desvpad muda dependendo se foi setado fonte de informação histórica
           • Cálculo do pedido
             • Input inicial
               • 1)calculatedLeadTimeAvg = leadTimeAverage; 2)leadTimeStandardDeviation = leadTimeDeviation; 3)leadTimeDemandDeviation = Math.Sqrt((calculatedLeadTimeAvg * Math.Pow(demandStandardDeviation(*s), 2)) + (Math.Pow(demandAverage, 2) * Math.Pow(leadTimeStandardDeviation, 2))); 4)leadTimeDemandAverage = demandAverage * calculatedLeadTimeAvg; 5)leadTimeDemandAverage = demandAverage * calculatedLeadTimeAvg; 6)annualDemand = demandAverage * 12;
               • aux = requestmark
               • requestMark = leadTimeDemandAverage + 1
               • expectedServiceLevel1= GeneralizedCompound2LeadTimeNonStockoutProbability
                 • GeneralizedCompound2LeadTimeNonStockoutProbability (calculatedDemandAvg, calculatedDemandDev, leadTimes, nonConsumptionProbability, requestMark);
               • expectedServiceLevel2 = GeneralizedCompound2AttendedDemandPercentage
                 • GeneralizedCompound2AttendedDemandPercentage(calculatedDemandAvg, calculatedDemandDev, leadTimes, nonConsumptionProbability, requestMark, lotsRestriction, expectedServiceLevel1);
                 • GeneralizedCompound2AttendedDemandPercentage
                   • -Inputs: average,deviation, leadTimes[][], nonConsumptionProbability, requestMark
                   • Variáveis iniciais
                     • leadTimeLength = leadTimes[1].Length
                     • result = 0.0
                   • Preenche o valor do desvio padrão em cada indice do vetor
                     • for (var l = 0; l < leadTimeLength; l++)
                     • aux = 0.0 & lt = leadTimes[0][l];
                     • for (var i = 1; i <= lt; i++)
                       • aux = aux + C(lt,i) * (1- nonConsumptionProbability)^i * nonConsumptionProbability^(lt - i) * deviation * Math.Sqrt(i) * normalLostValue
                       • standardDeviationsQty = (requestPoint - average * i) / (deviation * Sqrt(i));
                       • normalLostValue = MathHelpers.NormDist(standardDeviationsQty, 0, 1, false) + standardDeviationsQty * MathHelpers.NormSDist(standardDeviationsQty) - standardDeviationsQty;
                     • result += leadTimes[1][l] * aux;
                     • if (result < expectedServiceLevel1); result = expectedServiceLevel1;
             • Iterações para definir o requestMark otimizado
               • Ciclos de repetição igual a sequência do input inicial, contudo com loops de iteração onde altera por cada iteração o requestMark até atingir algum critério da iteração
               • Iteração 1: requestMark = requestMark/ 2
                 • while (expectedServiceLevel1 &gt; wantedP1 &amp;&amp; expectedServiceLevel2 &gt; wantedP2 &amp;&amp; requestMark &gt; 0)
               • Iteração Condicional 1: requestMark = requestMark + 1
                 • -if (requestMark.Equal(0)) while (expectedServiceLevel1 &lt; wantedP1 || expectedServiceLevel2 &lt; wantedP2)
                 • -Iteração condicional finaliza o processo caso chamada
                 • Finaliza
               • Iteração 2: requestMark = 2 * requestMark
               • Redefinição de variáveis
                 • requestMark= (requestMark/ 2) + 1);
                 • multiplier = 1
                 • expectedServiceLevel[1,2] = 0
               • Iteração 3: requestMark + (multiplier * aux)
                 • while (expectedServiceLevel1 &lt; wantedP1 || expectedServiceLevel2 &lt; wantedP2)
               • Condicional 2: Redefine variáveis e Iteração condicional
                 • Redefine variáveis
                   • requestMark = requestMark - (multiplier * aux)
                   • multiplier = multiplier / 2;
                   • expectedServiceLevel[1,2] = 0
                 • Iteração condicional 3: requestMark = requestMark + 1
                   • -if (multiplier * aux &lt; 1) while (expectedServiceLevel1 &lt; wantedP1 || expectedServiceLevel2 &lt; wantedP2)
           • Definição dos níveis
             • Cálculo do mínimo
               • min = requestMark + consumptionLot
               • Minimal limit
                 • -if (calculatedLeadTimeAvg &gt; 12)&nbsp; minimalLimit = (calculatedLeadTimeAvg / 12) * minimalLimit;&nbsp;
               • min = minimalLimit
               • Redefine LOS
                 • 1)expectedServiceLevel1 = GC2LNSO(**) a)if (expectedServiceLevel1 &lt; wantedP1) -expectedServiceLevel1 = wantedP1; 2)expectedServiceLevel2 = GC2ADP(**) a)if (expectedServiceLevel2 &lt; wantedP1)-expectedServiceLevel2 = wantedP2;
             • Cálculo do máximo
               • max = requestMark + lotsRestriction
                 • if (max &lt; min); max = min;
               • securityStock = min - consumptionLot - leadTimeDemandAverage;
               • averageStock = ((max - min + consumptionLot) / 2) + securityStock;
               • if (securityStock < 0) ; securityStock = 0; if (averageStock < 0) ; averageStock = 0;
             • resuppliesNumber = demandAverage * 12 / (max - min + consumptionLot);
           • Resultados finais e avaliações
       • Análise de resultados