Distribution Logistics Course

Introduction:

This comprehensive course presents a complete and balanced treatment of distribution logistics by covering both applications and the required theoretical background, therefore extending its reach to practitioners and students in a range of disciplines such as management, engineering, mathematics, and statistics. It serves as a useful reference for practitioners in the fields of applied mathematics and statistics, manufacturing engineering, business management, and operations research.

On completion of this course you should be able to: 

• Deal with the quantitative approaches needed to handle real-life management problems.
• Identify the limitations and scope of applicability of the proposed quantitative tools.
• Discuss many issues on probability and statistics as well as mathematical programming
• Gain a broad understanding on Network design and transportation, and Demand forecasting
• Examines inventory control in single- and multi-echelon systems, and Incentives in the supply chain
• Identify Network routing problems and develop solution methods for symmetric TSP.

Course Outline

Supply chain management

• What do we mean by logistics?

• Plan of the chapter.

•  Structure of production/distribution networks.
• Competition factors, cost drivers, and strategy.

• Competition factors.

• Cost drivers.
• Strategy.

• The role of inventories.

• A classical model: Economic Order Quantity.
• Cycle vs. capacity-induced stock.

• Dealing with uncertainty.

• Setting safety stocks.
• A two-stage decision process: Production planning in an assemble-to-order environment.

• Inventory deployment.

• Physical flows and transportation.
• Time horizons and hierarchical levels.
• Decision approaches.
• Information flows and decision rights.
• Quantitative models and methods.
• For further reading.

Network Design and Transportation

• The role of intermediate nodes in a distribution network.

• The risk pooling effect: reducing the uncertainty level.
• The role of transit points in transportation optimization.

• Location and flow optimization models.

• The transportation problem
• The minimum cost flow problem.
• The plant location problem
• Putting it all together

• Models involving nonlinear costs.
• For Further Reading.

Forecasting

• Overview on forecasting.
• The variable to be predicted.

• The forecasting process.

• Metrics for forecast errors.

• The Mean Error.
• Mean Absolute Deviation.
• Root Mean Square Error.
• Mean Percentage Error and Mean Absolute Percentage Error.
• ME%, MAD%, RMSE%.
• U Theil’s statistic.
• Using metrics of forecasting accuracy.

• A classification of forecasting methods
• Moving Average

• The demand model.
• The algorithm.
• Setting the parameters.
• Drawbacks and limitations.

• Simple exponential smoothing.

• The demand model.
• The algorithm.
• Setting the parameter.
• Initialization.
• Drawbacks and limitations.

• Exponential Smoothing with Trend.

• The demand model.
• The algorithm.
• Setting the parameters.
• Initialization.
• Drawbacks and limitations.

• Exponential smoothing with seasonality.

• The demand model.
• The algorithm.
• Setting the parameters.
• Initialization.
• Drawbacks and limitations.

• Smoothing with seasonality and trend.

• The demand model.
• The algorithm.
• Initialization.

• Simple linear regression.

• Setting up data for regression.

• Forecasting new products.

• The Delphi method and the committee process.
• Lancaster model: forecasting new products through products features.
• The early sales model.

• The Bass model.

• Limitations and drawbacks.

Inventory management with Deterministic Demand

• Economic Order Quantity.
• Robustness of EOQ model.
• Case of LT > 0: the (Q,R) model.
• Case of finite replenishment rate.
• Multi-item EOQ.

• The case of shared ordering costs.
• The multi-item case with a constraint on ordering capacity.

• Case of nonlinear costs.
• The case of variable demand with known variability.

Inventory control: the stochastic case.

• The newsvendor problem.

• Extensions of the Newsvendor problem.

• Multi-period problems.
• Fixed quantity: the (Q,R) model.

• Optimization of the (Q,R) model in case the stock out cost depends on the size of the stock out.
• (Q,R) system: case of constraint on the type II service level.
• Optimization of the (Q,R) model in case the cost of a stock-out depends on the occurrence of a stock out.(Q,R) system: case of constraint on type I service level.

• Periodic review: S and (s, S) policies.
• The S policy.
• The (s, S) policy.

Managing inventories in multiechelon supply chains

• Managing multi-echelon chains: Installation vs. Echelon Stock.

• Features of Installation and Echelon Stock logics.

• Coordination in the supply chain: the Bullwhip effect.
• A linear distribution chain with two echelons and certain demand.
• Arbores cent chain with two echelons: transit point with uncertain demand.
• A two echelon supply chain in case of stochastic demand.

Incentives in the supply chain

• Decisions on price: double marginalization.

• The first best solution: the vertically integrated firm.
• The vertically disintegrated case: independent manufacturer and retailer.
• A way out: designing incentive schemes.

• Decision on price in a competitive environment.

• The vertically disintegrated supply chain: independent manufacturer and retailer.

• Decision on inventories: the Newsvendor problem.

• The first best solution: the vertically integrated firm.
• The vertically disintegrated case: independent manufacturer and retailer.
• A way out: designing incentives and re-allocating decision rights.

• Decision on effort to produce and sell the product.

• The first best solution: the vertically integrated firm.
• The vertically disintegrated case: independent retailer and manufacturers.
• The case of efforts both at the upstream and downstream stage.

Vehicle Routing

• Network routing problems.
• Solution methods for symmetric TSP.

• Nearest-neighbor heuristic.
• Insertion-based heuristics.
• Local search methods.

• Solution methods for basic VRP.

• Constructive methods for VRP.
• Decomposition methods for VRP: cluster first, route second.

• Additional features of real-life VRP.

• Constructive methods for the VRP with time windows.