A Coflow structure to Screen Projects in a Product Pipeline

Informações

Código: GCT641
Divisão: GCT - Gestão de Ciência, Tecnologia e Inovação
Tema de Interesse: Tema 08 - Inovação e Gestão de Projetos

Autores

Paulo Soares Figueiredo, Elisabeth Loiola

Resumo

In many situations, System Dynamics modelers have to capture attributes of itemstracked in an aging chain by means of a co-flow. This study presents an application of coflowsin aging chains: A co-flow that enables the process of screening, i.e. the process ofeither terminating or approving projects. The article is the second in a series of two articlesabout a model to be presented at ENANPAD 2011 and describes a specific application for theprocess. An application to Product Pipeline Management (PPM) is developed. We model atwo-stage product development pipeline with a co-flow structure that tracks the number ofprojects and the related net present value (NPV) of payoff. Managers at each stage mustdecide on capacity utilization and thresholds for minimum value of projects. Simulationresults are shown and analyzed.The simplest configuration of the screening process in the NPD (New ProductDevelopment) context is the single-stage model with co-flow. This simplified configurationhas items (projects) that are initiated, developed and moved to the review stock, in whichthey are finally evaluated and either completed and taken to the next stage, or terminated.While projects are being developed, value creation is being added to the co-flow. Such valueaccumulates in the Value stock, and is either lost or transferred to the next stage together withthe corresponding projects.In order to know which part of the stock of items (projects) and of the stock of valuehas to be terminated, it is necessary to know the fraction of items that have a value lower thanthe predetermined threshold. In order to calculate such fractions, it is necessary to determinehow the population of values of items is distributed, i.e. what the probability distributionfunction (PDF) of values is. The choice of different thresholds will result in a differentfraction of projects that are accepted. A higher threshold will necessarily reduce such fraction.All the equations for the model can be found in the first paper of the series. Themanner in which our model has been set up differs from inventory/ service supply chainmodels (Sterman 1989, Anderson and Morrice 2005) both in terms of stock/flow and policystructures. The key structural difference is that inventory and service supply chain models donot usually have exit flows (aka screens).For the parameter set used in our tests, the PPM model does not exhibit any bullwhipwhen the screens are in place. However, under extreme conditions (Forrester and Senge1980), when the screens are eliminated (thresholds=0), together with a bias towards reducingbacklogs and a specific choice of nominal development times and time to adjust capacity, themodel reverts back to a bullwhip effect similar to that found in a service supply chain.Simulation results presented here illustrate that screening can eliminate the backlogbullwhip effect in the pipeline, and that policies that minimize “backlog bullwhip” may not bethe best policies to minimize “value bullwhip”.

Abrir PDF