Supply Chain Simulation: A System Dynamics Approach for Improving Performance (review)

In 1958 Jay Forrester presented DYNAMO, one of the early simulation languages specifically developed for modeling complex dynamic systems, providing the most seminal contribution to understanding the bullwhip phenomenon (see J. Forrester, “Industrial Dynamics: A Major Breakthrough for Decision-makers,” Harvard Business Review 36 [1958]: 37–66). Beside this outcome, Forrester’s work is outstanding for a further reason: it sets out the field of supply chain simulation. Since then Forrester’s approach, commonly known as system dynamics, has largely contributed to the development and improvement of supply chain analysis. In this book, Francisco Campuzano and Josefa Mula present a practical guide to supply chain modeling and simulation via system dynamics. The book is organized into seven chapters.

Chapters 1–4 focus on theoretical principles of supply chain simulations. Chapter 1 reviews and briefly discusses the features of the simulation approaches used to study the dynamics of supply chains. Four methods are discussed: spreadsheet-based simulation, system dynamics, discrete events systems, and dynamics and business games. In chapter 2, the authors resume the three fundamental interrelated elements of supply chains: network, business processes, and components. Chapter 3 focuses on the bullwhip effect. First, the authors describe classical examples of demand amplification phenomenon, for example, the beer game. Furthermore they resume the four main causes of this detrimental phenomenon and classify the related avoiding strategies. Additionally they examine two performance metrics for assessing the bullwhip effect: the order rate variance ratio and the net stock amplification. Finally they review the performance of four supply chain strategies: traditional, e-shopping, vendor managed inventory, and electronic point of sale (EPOS). Chapter 4 presents the main concepts of the system dynamics methodological approach. First, a detailed explanation of how to build a generic system dynamic model is presented. Finally, they report a description of the specific steps to develop a supply chain simulation model. The strength of this chapter lies in the identification of the variables that should be considered to model a supply chain structure.

Chapters 5–7 provide a practical support for the generation of supply chain models through Vensim, a system dynamics software package. In chapter 5, the authors introduce the causal loop diagram and the Forrester’s diagram of a single stage supply chain. Thus, they show how to simulate warehouse models under different assumptions. More specifically three examples are formalized. Problem 1 models a single-stage supply chain in which order backlogging is not allowed. In problem 2, the authors include the backlog assumption. Finally, problem 3 illustrates how to model a single-stage supply chain with perishable products. Chapter 6 presents the causal loop diagram of a three-stage traditional supply chain in which the trading partners adopt the APIOBPCS, a smoothing (R, S) inventory control policy. The authors identify and describe each variable and parameter to be used for modeling and simulation of a multi-echelon supply chain. In chapter 7 the authors present a comprehensive traditional supply chain model. Vensim is adopted to create the Forrester’s diagram based on the causal loop diagram described in the previous chapter. The entire code for each variable of the model is reported and explained.

A benefit that comes from this book is related to its relatively easy-to-follow explanations. System dynamics concepts for modeling, studying, and analyzing different supply chain strategies are presented in a didactic and communicative manner. The strength of this book lies in the second part in which the authors illustrate the generation of the causal-loop diagrams and Vensim instructions and equations that have to be used to model relevant assumptions of the production-inventory system, such as the no-negative condition of the inventory, the forecast demand techniques, and the manufacturer lead time. It is important to point out that even if practical applications are detailed to be developed in Vensim, this knowledge can also be applied to generate a supply chain simulation model through...