SC0x - Supply Chain Analytics, is delivered on a learner-paced schedule. This course contains the mathematical concepts used throughout all the other courses and is packaged in modules based on content. It is the reference course for all the others and so remains available as needed almost continually. Each run of SC0x will have a final exam on a date set by the course team. The course will close briefly after each scheduled final exam and learners can enroll in the next run of the course if needed. All course content, practice problems, exams, reading materials, and external links are delivered through the course platform on edX. In SC0x there is one final exam that will count for 100% of your grade.
INTRODUCTION TO SUPPLY CHAINS AND BASIC ANALYSIS: In this first week, we will provide an overview of supply chains. We will introduce some of the basic concepts and approaches of the discipline. We will also provide a review of the basics of analysis: models, algebra and mathematical functions. You will learn how to use descriptive, predictive, and prescriptive models in making supply chain decisions.
PRESCRIPTIVE MODELING I: CONSTRAINED AND UNCONSTRAINED OPTIMIZATION: In this week we introduce optimization in the form of prescriptive models. You will learn when and how to use classic optimization techniques to find the minimum or maximum values of an unconstrained cost or profit function. You will also be introduced to linear programs (LPs) for constrained problems. LPs are the most commonly used models for decision making in supply chains. At the end of this week you should be very comfortable with how to formulate and solve LPs for a wide variety of different problem types.
PRESCRIPTIVE MODELING II: IPS, MILPS, AND NETWORK MODELS: In this week we extend our discussion of constrained optimization to include integer programming (IP), mixed integer linear programming (MILP), and network models. You will learn how to use Integer and Binary variables to better represent real-life supply chain decisions, such as facility location, network design, production planning, etc. You will become adept at formulating, solving, and interpreting the results of LPs, IPs, and MILPs as well as gaining an understanding of basic network optimization methods to solve shortest path, traveling salesman, and vehicle routing problems.
ALGORITHMS AND APPROXIMATIONS: In this week you will learn two approaches to problem-solving that are very common in supply chain management: algorithms and approximations. Algorithms are everywhere! You will learn the basic concepts of developing and deploying algorithms and will see how they are used in some fundamental supply chain applications, such as vehicle routing and inventory planning. Sometimes, however, exact solutions are unfeasible or unobtainable within the desired time. In these cases, approximation methods can be applied quite successfully. We will discuss and go into some detail on one approximation method for estimating costs of vehicle routing to illustrate the approach.
MANAGING UNCERTAINTY: DISTRIBUTIONS AND PROBABILITY: In this week, we introduce the added complexity of uncertainty (or stochasticity) to our analysis. You will learn how to measure, model, and manage uncertainty and randomness within supply chains. You will become comfortable with a variety of continuous and discrete probability distributions that are widely used in supply chains, such as Normal, Uniform, Poisson, and others.
STATISTICAL TESTING: This week is all about the statistical testing. You will learn how to conduct hypothesis testing. You will learn how to formulate, test, and analyze the results of various forms of tests that are widely used in practice.
REGRESSION AND SIMULATION MODELS: Often times you will need to develop a model to better predict the future in order to improve planning. In this week you will learn how to develop econometric models, mainly ordinary least squares (OLS) linear regression, that use past history to better estimate the future. OLS regression is widely used to estimate future product demand as well as to better understand how different independent factors influence a dependent variable. In this week, you will not only learn how to build these econometric models, but also how to interpret and use them in practical scenarios. Regression is a predictive model that measures the impact of independent variables on dependent variables. We then cover simulation, which captures the outcomes of different policies with an uncertain or stochastic environment.
QUEUEING THEORY AND DISCRETE EVENT SIMULATION: In this week you will learn the basics of queueing theory to help you better manage supply chain processes. Supply chains can be quite complex. The impact of one firm's decisions on another firm's costs, for example, are not always known ahead of time. In order to better describe these complex situations, we often use simulation models. Simulation helps us understand what an outcome will be given a set of inputs. It is very different from prescriptive (optimization) models that makes recommendations. In this week you will learn how to use and interpret the results of discrete event and Monte Carlo simulations.
FINAL EXAM: The final exam is a timed exam that covers all material from the course. See course for exam instructions.