Forecasting Workshop: Intermittent Demand Forecasting

Similar documents
A Meta Forecasting Methodology for Large Scale Inventory Systems with Intermittent Demand

THINK DEVISE CALCULATE ENVISION INVENT INSPIRE THINK DEVISE CALCULATE ENVISION INV

Intermittent Demand Forecasts with Neural Networks

Forecasting irregular demand for spare parts inventory

The Integrated Inventory Management with Forecast System

Managing the demand for spare parts

CHAPTER 3 CALL CENTER QUEUING MODEL WITH LOGNORMAL SERVICE TIME DISTRIBUTION

FORECASTING IN SAP-SCM (SUPPLY CHAIN MANAGEMENT)

Optimal policies for demand forecasting and inventory management of goods with intermittent demand

The accuracy of intermittent demand estimates

Product Documentation SAP Business ByDesign Supply Chain Planning and Control

Optimizing Inventory in Today s Challenging Environment Maximo Monday August 11, 2008

Forecasting Aviation Spare Parts Demand Using Croston Based Methods and Artificial Neural Networks

Multi-item Sales Forecasting with. Total and Split Exponential Smoothing

Forecasting and big data analysis

Evaluating the Lead Time Demand Distribution for (r, Q) Policies Under Intermittent Demand

FORECASTING FOR THE ORDERING AND STOCK- HOLDING OF CONSUMABLE SPARE PARTS

Sales and operations planning (SOP) Demand forecasting

DYNAMIC FUZZY PATTERN RECOGNITION WITH APPLICATIONS TO FINANCE AND ENGINEERING LARISA ANGSTENBERGER

Electronic Audit Evidence (EAE) and Application Controls. Tulsa ISACA Chapter December 11, 2014

forecast modeling Statistical Calculations, Replacement Items, and Forecast Templates for Demand Planning Industries Required Modules:

Keep the balance with 8 inventory management viewpoints

Strengthening Diverse Retail Business Processes with Forecasting: Practical Application of Forecasting Across the Retail Enterprise

Anytime 500 Forecast Modeling

An Introduction to SAS Enterprise Miner and SAS Forecast Server. André de Waal, Ph.D. Analytical Consultant

Integrated Business Planning Overview and Preview of the New Demand Application. Tod Stenger SAP

Dan French Founder & CEO, Consider Solutions

Applying Actual Usage Inventory Management Best Practice in a Health Care Supply Chain

A Case Study Analysis of Inventory Cost and Practices for Operating Room Medical/Surgical Items

EMIR Key business impacts for asset managers

Predictive Analytics

Role of Forecast Analytics in Inventory Management

Use of Statistical Forecasting Methods to Improve Demand Planning

SAP Predictive Analytics Roadmap Charles Gadalla SAP SESSION CODE: #####

Forecasting method selection in a global supply chain

Marketing Mix Modelling and Big Data P. M Cain

Pentaho Data Mining Last Modified on January 22, 2007

Identification of Demand through Statistical Distribution Modeling for Improved Demand Forecasting

Development Period Observed Payments

Text Mining in JMP with R Andrew T. Karl, Senior Management Consultant, Adsurgo LLC Heath Rushing, Principal Consultant and Co-Founder, Adsurgo LLC

Auto Days 2011 Predictive Analytics in Auto Finance

Ch.3 Demand Forecasting.

Operative Media. Operative.One. Release Finance Module User's Guide. Document Version 1.0

2015 MATLAB Conference Perth 21 st May 2015 Nicholas Brown. Deploying Electricity Load Forecasts on MATLAB Production Server.

SAP Sales and Operations Planning Compare & Contrast with SAP APO. Phil Gwynne SAP UKI 2013

Master Data Management Enterprise Architecture IT Strategy and Governance

SUPPORTING LOGISTICS DECISIONS BY USING COST AND PERFORMANCE MANAGEMENT TOOLS. Zoltán BOKOR. Abstract. 1. Introduction

Models for Product Demand Forecasting with the Use of Judgmental Adjustments to Statistical Forecasts

Solution-Driven Integrated Learning Paths. Make the Most of Your Educational Experience. Live Learning Center

Practical Time Series Analysis Using SAS

SaPHAL Sales Prediction powered by HANA and Predictive Analytics

Improving Demand Forecasting

4. Simple regression. QBUS6840 Predictive Analytics.

A case study on volatile demand

VDI FIT and VDI UX: Composite Metrics Track Good, Fair, Poor Desktop Performance

Forecasting, Prediction Models, and Times Series Analysis with Oracle Business Intelligence and Analytics. Rittman Mead BI Forum 2013

Operational Analytics for APO, powered by SAP HANA. Eric Simonson Solution Management SAP Labs

Advanced Data Management and Analytics for Automated Demand Response (ADR) based on NoSQL

How Organisations Are Using Data Mining Techniques To Gain a Competitive Advantage John Spooner SAS UK

FORECASTING METHODS FOR SPARE PARTS DEMAND

Self-Adaptive Workload Classification and Forecasting for Proactive Resource Provisioning

Data Collection Database options and Data Management considerations. Paul Donnelly George Clinical The George Institute for Global Health

USE OF ARIMA TIME SERIES AND REGRESSORS TO FORECAST THE SALE OF ELECTRICITY

Experiences with Online Programming Examinations

2015 Workshops for Professors

Automating FP&A Analytics Using SAP Visual Intelligence and Predictive Analysis

Technical Indicators Tutorial - Forex Trading, Currency Forecast, FX Trading Signal, Forex Training Cour...

Data Mining Practical Machine Learning Tools and Techniques

Plan forecast optimise

Managing Complex Capital Portfolio using PPM integration with PS, CATS & FICO. Session: Sanjaya Una

Improving Software Quality with the Continuous Integration Server Hudson. Dr. Ullrich Hafner Avaloq Evolution AG 8911

Leveraging Ensemble Models in SAS Enterprise Miner

Industry Environment and Concepts for Forecasting 1

SAP APO SNP (Supply Network Planning) Sample training content and overview

Predictive Analytics Powered by SAP HANA. Cary Bourgeois Principal Solution Advisor Platform and Analytics

MARKETING AUTOMATION TRENDS

COMBINING THE METHODS OF FORECASTING AND DECISION-MAKING TO OPTIMISE THE FINANCIAL PERFORMANCE OF SMALL ENTERPRISES

THE USE OF PREDICTIVE MODELLING TO BOOST DEBT COLLECTION EFFICIENCY

Azure Machine Learning, SQL Data Mining and R

Data Mining mit der JMSL Numerical Library for Java Applications

QSC AG. Company Presentation. Results Q Cologne, 10 August 2015

Moving averages. Rob J Hyndman. November 8, 2009

Automating IT Capacity Management

USING FORECASTING TOOLS

Transcription:

Workshop: Intermittent Demand Software and Freeware for Intermittent Demand Nikolaos Kourentzes nikolaos@kourentzes.com

Introduction Research over the last years has developed multiple methods and innovations in intermittent/slow moving items forecasting. A key barrier to using these in practice is their availability in mainstream software. In this presentation we will present some of the available options and potential solutions. We will look at two sides of the question: Examples of available commercial software; A free (GPL) toolbox for intermittent demand for R. This is not a complete list and the purpose of this presentation is not to provide one, but rather to illustrate some of the options and issues to consider.

Agenda Agenda 1. Commercial software 2. Non-commercial software 3. Demo

Introduction We will look at three aspects: 1. Availability of methods/models and transparency of implementation 2. Automation in terms of forecasting method selection and setup 3. Support for exploration and analysis of intermittent time series

Commercial Typically intermittent demand forecasting methods are included in demand planning and specialised forecasting software. We will look at the following examples: SAP: Advanced Planning and Optimisation (APO) ForecastPro (TRAC edition) SAS Forecast Server A notable mention is Smart Software: Intermittent Demand Planning and Service Parts, which is focused at using a bootstrapping based forecasting approach.

Commercial: SAP APO SAP APO offers limited intermittent demand forecasting. It offers a single method: Croston s method (original): Fixed model initialisation (maybe rather problematic) Manual selection of (a single) method parameter Two ways to get a forecast out of the method: constant and sporadic Sporadic takes the last interval and demand estimates and assumes some deterministic behaviour thereafter. Do not use! Constant is the standard Croston s forecast

Commercial: SAP APO SAP APO offers limited exploration and time series analysis facilities. This is true for the intermittent demand case. SAP APO offers some model selection between continuous demand forecasting methods and Croston s. Time series that have 66% or more of their history zero demand periods are assigned to Croston s. This threshold can be adjusted, but SAP APO does not offer any analytics tools to help decide an appropriate threshold. The details of the implementation are documented online: http://help.sap.com/saphelp_scm50/helpdata/en/ac/216b89337b11d398290000e8a4960 8/content.htm

Commercial: ForecastPro ForecastPro is a dedicated forecasting software with limited demand planning support. It offers two options based on Croston s method: Original Croston s method Modification of Croston s as suggested by Willemain et al. (assumes log-normal distribution of order sizes instead of normal) ForecastPro always uses Willemain et al. modification, unless there are negative values in the history, when it reverts to Croston s original.

Commercial: ForecastPro ForecastPro automatically decides when to use Croston s method over continuous demand methods/models. The choice rule is unknown. The user can manually override the method selection but there are no exploration tools to support the choice. Method parameters are chosen automatically (unknown methodology). Demand size and intervals have separate smoothing parameters, giving more flexibility to the model in accordance to current research. No option for manual parameters.

Commercial: ForecastPro ForecastPro is a dedicated forecasting software with limited demand planning support. It offers two options based on Croston s method: Original Croston s method Modification of Croston s as suggested by Willemain et al. (assumes log-normal distribution of order sizes instead of normal) ForecastPro always uses Willemain et al. modification, unless there are negative values in the history, when it reverts to Croston s original.

Commercial: ForecastPro

Commercial: SAS Forecast Server SAS Forecast Server stands in the middle of pure forecasting software and demand planning/inventory management software. It does offer some support for intermittent time series forecasting. Two methods are implemented: Croston s method type Average demand forecast SAS has implemented Croston s time series decomposition, but within its suite there is no restriction to the model used to predict either demand intervals or size.

Commercial: SAS Forecast Server Although this raises the flexibility of the method it raises questions of model/parameter selection, thus complicating the model setup and automation. SAS Forecast Server does not offer additional tools to explore the intermittent time series further to help decide on methods and parameter choice. It is perhaps useful to keep in mind that SAS permits easily to code new functions and as such the current capabilities can be substantially extended in-house.

Some common themes for the various commercial software: Intermittent Demand Software Commercial: Overview Mostly focus on Croston s method, typically ignore developments on bias, obsolesce corrections and lack of independence between demand events and demand size. Offer very limited tools to explore and understand intermittent data. Opaque parameter selection Nonetheless implemented within a forecasting system (hierarchies, adjustments, etc). In no way an exhaustive list!

Agenda Agenda 1. Commercial software 2. Non-commercial software 3. Demo

The lack of thorough and standardised implementations of intermittent demand methods is not exclusive to commercial software. This has lead to develop a freely available and open source (GPL 2) intermittent demand package for R. Intermittent Demand Software Non-commercial The motivation in developing this was to have a standardised implementation of methods, following current research findings, that is transparent and open source for others to use and build on. The package is freely available at either CRAN R repository or my research blog http://cran.r-project.org/web/packages/tsintermittent/index.html

tsintermittent R package Exploration and method selection Conventional plots: Time series plots Histograms of demand Intermittent demand specific plots: Non-zero demand size and interval plot (classification of demand) Classification of demand into: erratic, lumpy, intermittent and smooth. This classification has been beneficial for method selection, between Croston s original method and the bias-corrected SBA.

tsintermittent R package Exploration and method selection Multiple classification have been proposed, with an objective of method selection: Syntetos et al., 2005 Kostenko and Hyndman, 2006 Petropoulos and Kourentzes, 2014 These model selection schemes have lead to improvements in forecasting performance

Demand size Intermittent Demand Software tsintermittent R package Exploration and method selection... but also permit a more detailed understanding of the intermittent items in the assortment: High demand variance, low intermittence SKU. High intermittence, low variance SKU. Average demand interval

tsintermittent R package Methods tsintermittent package offers a variety of methods (which is expanded over time): Croston s and variants: Croston s original method SBA and SBJ modifications Croston decomposition based moving averages (with bias correction) TSB method for obsolescence Temporal aggregation methodologies: Aggregate-Disaggregate Intermittent Demand Approach (ADIDA) Intermittent Multiple Aggregation Prediction Algorithm (imapa) Single exponential smoothing

tsintermittent R package Methods Croston s method options: Manual or automatic parameter selection based on latest research Single or independent smoothing parameters for demand size and intervals Modifications (SBA, SBJ) Similar Croston based moving average and TSB method options.

tsintermittent R package Methods Single exponential smoothing: Intermittent demand series consistent automatic parameter selection Conventional parameter optimisation Intermittent series specific optimisation

tsintermittent R package Methods Temporal aggregation methods: model the intermittent time series at larger time buckets to reduce intermittency: ADIDA imapa A: Original data (months) B: Aggregate data (quarters) C: A quarterly forecast is produced D: The quarterly forecast is broken down to three monthly forecasts

tsintermittent: Overview Free open source package for intermittent demand time series analysis and forecasting. Updated with latest research and continuously expanding list of methods and tools. Not implemented within a demand planning system, but several options to incorporate R modules in existing IT infrastructure. Useful links: N. Kourentzes and F. Petropoulos, 2014, tsintermittent R package: http://cran.r-project.org/web/packages/tsintermittent/index.html Blog post outlining functionality of tsintermittent: http://kourentzes.com/forecasting/2014/06/11/on-intermittent-demand-modeloptimisation-and-selection/ Online package demo: https://kourentzes.shinyapps.io/shinyintermittent/

Demonstration of tsintermittent https://kourentzes.shinyapps.io/shinyintermittent/

Thank you for your attention! Questions? Nikolaos Kourentzes Lancaster University Management School - Lancaster, LA1 4YX email: nikolaos@kourentzes.com blog: http://nikolaos.kourentzes.com / Full or partial reproduction of the slides is not permitted without author s consent. Please contact n.kourentzes@lancaster.ac.uk for more information.