Transforming big data into supply chain analytics



Similar documents
2 Day In House Demand Planning & Forecasting Training Outline

Enhancing Sales and Operations Planning with Forecasting Analytics and Business Intelligence WHITE PAPER

Enhancing Sales and Operations Planning with Forecasting Analytics and Business Intelligence WHITE PAPER

Business Intelligence & Product Analytics

ElegantJ BI. White Paper. Key Performance Indicators (KPI) A Critical Component of Enterprise Business Intelligence (BI)

Planning Demand For Profit-Driven Supply Chains

Data Mining for Successful Healthcare Organizations

QAD Business Intelligence

Make the right decisions with Distribution Intelligence

Alexander Nikov. 5. Database Systems and Managing Data Resources. Learning Objectives. RR Donnelley Tries to Master Its Data

SQL Server 2005 Features Comparison

IT S ALL ABOUT THE CUSTOMER FORECASTING 101

Vendor briefing Business Intelligence and Analytics Platforms Gartner 15 capabilities

Foundations of Business Intelligence: Databases and Information Management

Course 6234A: Implementing and Maintaining Microsoft SQL Server 2008 Analysis Services

Dashboard Reporting Business Intelligence

Implementing Data Models and Reports with Microsoft SQL Server

Fluency With Information Technology CSE100/IMT100

Business Intelligence Solutions. Cognos BI 8. by Adis Terzić

Enterprise Solutions. Data Warehouse & Business Intelligence Chapter-8

OLAP Theory-English version

Course MIS. Foundations of Business Intelligence

Chapter 6 FOUNDATIONS OF BUSINESS INTELLIGENCE: DATABASES AND INFORMATION MANAGEMENT Learning Objectives

Foundations of Business Intelligence: Databases and Information Management

End to End Microsoft BI with SQL 2008 R2 and SharePoint 2010

CHAPTER SIX DATA. Business Intelligence The McGraw-Hill Companies, All Rights Reserved

Chapter 6. Foundations of Business Intelligence: Databases and Information Management

OLAP. Business Intelligence OLAP definition & application Multidimensional data representation

Enterprise Resource Planning Analysis of Business Intelligence & Emergence of Mining Objects

Foundations of Business Intelligence: Databases and Information Management

IT and CRM A basic CRM model Data source & gathering system Database system Data warehouse Information delivery system Information users

MS 50511A The Microsoft Business Intelligence 2010 Stack

Data Warehousing and Data Mining in Business Applications

QAD BUSINESS INTELLIGENCE

Business Analytics and Data Visualization. Decision Support Systems Chattrakul Sombattheera

Data Isn't Everything

Business Intelligence Osvaldo Maysonet VP Marketing & Customer Knowledge Banco Popular

Transforming Internal Audit: A Maturity Model from Data Analytics to Continuous Assurance

SQL Server 2012 End-to-End Business Intelligence Workshop

Key Performance Indicators used in ERP performance measurement applications

The metrics that matter

MSD Supply Chain Programme Strategy Workshop

Adding insight to audit Transforming internal audit through data analytics

Business Intelligence and Healthcare

msd medical stores department Operations and Sales Planning (O&SP) Process Document

Self-Service Business Intelligence: The hunt for real insights in hidden knowledge Whitepaper

Making the right decisions with SCOR

ProClarity Analytics Family

Business Intelligence, Analytics & Reporting: Glossary of Terms

CHAPTER - 5 CONCLUSIONS / IMP. FINDINGS

Chapter 6 8/12/2015. Foundations of Business Intelligence: Databases and Information Management. Problem:

Tapping the benefits of business analytics and optimization

Foundations of Business Intelligence: Databases and Information Management

The Microsoft Business Intelligence 2010 Stack Course 50511A; 5 Days, Instructor-led

AV TSS-05 Avantis.DSS 5.0 For Wonderware Intelligence

Database Marketing, Business Intelligence and Knowledge Discovery

Part 22. Data Warehousing

Seamless Dynamic Web Reporting with SAS D.J. Penix, Pinnacle Solutions, Indianapolis, IN

It s about you What is performance analysis/business intelligence analytics? What is the role of the Performance Analyst?

Integrated Sales and Operations Business Planning for Chemicals

Corporate Performance Management Framework

3. Provide the capacity to analyse and report on priority business questions within the scope of the master datasets;

Making Business Intelligence Relevant for Mid-sized Companies. Improving Business Results through Performance Management

Hexaware E-book on Predictive Analytics

Data Warehouse design

How To Model Data For Business Intelligence (Bi)

Data collection architecture for Big Data

Foundations of Business Intelligence: Databases and Information Management

Foundations of Business Intelligence: Databases and Information Management

LEARNING SOLUTIONS website milner.com/learning phone

Data Analytics in Organisations and Business

PRONTO-Xi Business Intelligence

Google AdWords, 248 Google Analytics tools, 248 GoogleAdsExtract.xlsx file, 161 GoogleAnalytics, 161

Supply Chain Optimization for Logistics Service Providers. White Paper

Social Business Intelligence For Retail Industry

{Businesss. Intelligence. Overview. Dashboard Manager

SQL Server Administrator Introduction - 3 Days Objectives

Licenze Microsoft SQL Server 2005

Data Management Practices for Intelligent Asset Management in a Public Water Utility

Selection Requirements for Business Activity Monitoring Tools

Optimizing the Source to Contract Process to Maximize and Lock in Savings Patrick Eckhert Cardinal Health Head of Indirect Procurement

Value of. Clinical and Business Data Analytics for. Healthcare Payers NOUS INFOSYSTEMS LEVERAGING INTELLECT

Management Accountants and IT Professionals providing Better Information = BI = Business Intelligence. Peter Simons peter.simons@cimaglobal.

Business Intelligence for Dynamics GP. Presented By: Rob Jackson, Business Intelligence Consultant Brent Keilin, GP Consultant

BI4Dynamics provides rich business intelligence capabilities to companies of all sizes and industries. From the first day on you can analyse your

September 17, 1:00 PM. Dean Sorensen, Founder, IBP Collaborative

Hospitality with a system. Web-based management tool. protel Business Intelligence. Product information. protel hotelsoftware GmbH

Implementing Oracle BI Applications during an ERP Upgrade

Advancing Your Business Analysis Career Intermediate and Senior Role Descriptions

Microsoft Implementing Data Models and Reports with Microsoft SQL Server

Using data analytics and continuous auditing for effective risk management

IBM Cognos Performance Management Solutions for Oracle

Transcription:

Transforming big data into supply chain analytics ALAN MILLIKEN CFPIM CSCP CPF CSOP Introduction Analytics has been described as finding and using meaningful information in big data to improve business performance. Today s information technology systems gather and store a tremendous amount of supply chain related data. To take advantage of this capability firms must transform data to business intelligence including analytics. In supply chain the ultimate goal is to convert the mass of unstructured data into useful analytics that help to improve service, reduce costs, improve inventory management and increase profits. In a 2012 SAS-MIT survey with 2,500 respondents from over 20 industries, 67% indicated they are using analytics to improve overall performance. Data mining, the process of extracting information from a data set and transforming it into a usable structure, supports analytics. It can be fully automatic using algorithms supported by advanced statistics, math and software programs or the mining process can be interactive driven by the end user. For example, online analytical processing (OLAP) of multi-dimensional data cubes (e.g. customer, location, sales) is integrated into advanced planning software to enable reporting, and support aggregation, drill-down and slicing & dicing of the data. Operationally users can develop their own custom analytics. For example, deploying end user defined filters or rules to find exceptions to a given rule. The data mining tool may be programmed to do cluster analysis, detect anomalies in the data, or apply association rules. Both analytics and data mining are growing as buzzwords that are used to describe any large scale gathering or analysis of data. This article will focus on the more narrow definition of these terms in supply chain management. The concept explained in real life There is widely available literature that explains in depth all the technical and intricate details relating to supply chain. For the novice and beginner, we will use a classical and authentic South African example to explain in broad terms and draw an analogy to the basic concepts. A braai is an Afrikaans work that refers to a grill or a barbecue. The braai is so popular in South Africa that there is even a national braai day in September. Let us consider an event where you decide to that you are going to organise a braai on a particular day. What are the sequence of events that must be taken into account and managed to host a successful braai? Managing those events and processes is simply supply chain. Approaches to use of analytics According to the MIT-SAS research about 10% of firms surveyed have become Analytic Innovators who leverage advanced analytics to re-think the business and innovate processes and products. About 60% have progressed to becoming Analytic Practitioners and 30% are still analytically challenged. In supply chain analytic practitioners use the information gained to solve problems, improve efficiencies, increase service and reduce inventories. The types of analytics used in supply chain management include: Descriptive analytics (e.g. reports, KPI s, dashboards) to report performance and determine what happened, why it happened and plan for change. SAPICS 2015, www.sapics.org.za ISBN 978-0-620-64684-0 PAGE 1

Operational level reports based on pre-determined querying logic models and end user specified queries to improve decisions and identify the need for action. Predictive analytics to improve such processes as forecasting, customer relationship management and inventory control. Basic decision models that use decision logic or business rules to help optimize or maximize outputs. Big data Terms used to refer to the mass of information being generated today. In 2012, it was estimated that 2.5 exabytes of data were created each day. (1 exabyte = 1B gigabytes) The amount of data available is expected to double every 3 years. Technology increases data availability, enables communication of data and provides the ability to analyse the information. Those firm who successfully transform this mass of information into analytics that can be used to make better decisions and act in a timely manner will gain a competitive sustainable advantage. Gathering and structuring data for analysis As George Bernard Shaw once said, Take care to get what you like, or you will be forced to like what you get! The information infra-structure and data structure must be designed to support data mining, reporting and analysis. For example, if developing a structure to support sales and forecasting analyses key characteristics must be included. Material dimensions SKU, Product, Product Group, etc. Customer dimensions Sold-To, Payer, Customer Group, etc. Accounting dimensions BU, SBU, Profit Center, etc. Geographical dimensions country, region, sub-region, state, customer, etc. Of course existing reports and analytics are a source of input but it would be a mistake to assume current information represents total need. Key users, process experts, external benchmarks, etc.., should be included in determining what information is required and how it will be used. It is better to err on the side of including too much information when designing the structure. In addition to determining what data elements are required, key figures must be defined based on end users needs. For example orders last year for the month, statistical forecast, etc... Key figures include historical data, data generated by the system and collection of qualitative inputs. At this point, the firm must also define what system generated analytics are desired. On-line analytical processing (OLAP) of such analytics is often integrated into the system and must be considered when structuring the data. For example what performance metrics will be generated routinely? Such things as forecast error and BIAS. Multi-dimensional data cubes provide key figures to support data mining. For example the cube below contains: Absolute values & quantities Sales by BU-Product-Region Forecast by SKU-Customer SAPICS 2015, www.sapics.org.za ISBN 978-0-620-64684-0 PAGE 2

Descriptive analytics These are used to measure performance, report what happened, why it happened and plan for improvement. Dashboards, Key Performance Indicators (KPI s) and diagnostics are common examples. These include such things as past performance measurement, quantitative analysis and qualitative inputs. The goal is to use this input to improve performance and decisions. Exception-based analysis are helpful to focus on current problems. Predictive analytics The analysis of current and/or historical data to make predictions about the future. These are used to improve performance in planning and controlling. For example, improved forecasting can increase service, reduce inventory and decrease costs simultaneously. Predictive analytics can be as simple as a 3-month moving forecast model. SAPICS 2015, www.sapics.org.za ISBN 978-0-620-64684-0 PAGE 3

Descriptive analytics, for example forecast accuracy measures are often used to improve the performance of predictive analytics, for example the forecast. The trend in supply chain management today is toward more complex predictive analytics to handle multiple variables that influence outcomes. For example, the analytic model below was created to predict significant downturns and upturns in product group demand at least 3 months before it occurs. Such analytics ensure an early and pro-active approach to managing the supply chain. Using analytics to optimize outcomes Because of the tremendous growth in data available and increased power to gather and process this data, the use of optimizers has become common. For example, assume a firm has four production plants located in four geographical regions. All or most products can be produced at all plants. The firm first develops an objective to drive the optimization process. For example, Maximize EBIT (Earnings before Interest and Taxes) within capacity constraints and inventory limits. The required inputs are defined and the software is programmed to produce outputs meeting the objective. Notice the system uses a combination of master data from the planning system, price/cost data from finance and the demand forecast (predictive analytic) to perform the optimization. As part of the output, sales forecast are assigned to customers and production volumes including specific products are assigned to plants. Distribution and inventory plans are generated to help plan logistics resources. Projected revenues and associated profits are key outputs. The ability to focus on multiple key variables thus ensuring the best overall plan has been made much easier through advanced data gathering and analysis tools. Summary The key steps to leveraging big data to become more competitive are: Identify what information is needed to make better decisions and take timely action. Review current analyses and talk with stakeholders. Develop a list of questions that need to be answered. Remember it is better to create too much data than not enough. Develop the data structure and reporting needed to provide the desired information. In the case of supply chain planning for example, material, geographical and business dimensions must be defined to support end user analysis. Design and implement the reporting and data query system for descriptive analytics. Standard output should include key performance indicators (e.g. Forecast Error, BIAS, etc.), exception reports (e.g. Sales with No Forecast, Forecast with No Sales, etc.), and standard activity reports (e.g. Sales by Customer). Design and implement the system for predictive analytics such as the demand forecast or an optimized production plan. Advanced software platforms are available that will perform these tasks but the firm still must identify what information they need and work with the consultant to identify and implement the process. Using big data to improve performance is a journey not a destination. The firm should have a process supported by people and technology to ensure information needs are continuously assessed and analytics are used to make the firm better. Use of analytics should be integrated into education & training programs and periodic audits performed to verify use of the systems. SAPICS 2015, www.sapics.org.za ISBN 978-0-620-64684-0 PAGE 4

In explaining why poor decisions were made based in part on inaccurate or incomplete information, former Secretary of Defense Donald Rumsfeld said, Sometimes you don t know what you don t know and in the same speech Sometimes you do know what you don t know. To maximize the benefits of big data and the associated analyses firms must determine what they need to know, create information flow aligned with their decision needs and take actions to ensure everyone in the organization leverages the information. SPEAKER PROFILE Alan Milliken is a Senior Manager on the Supply Chain Capability Development Team at BASF, the world s leading chemical company. He has extensive experience in manufacturing operations including supply chain management, as a supply chain consultant and a supply chain educator. He serves on the Board of Advisors at the Institute of Business Forecasting (IBF) and as a Subject Matter Expert he helped to create the Certified Professional Forecaster (CPF) program. Alan is a frequent speaker at supply chain events and has been published many times. He holds an engineering degree from Auburn University and an MBA from Clemson University. Contact details Email address alan.miliken@basf.com Website www.basf.co.za Telephone +018509371808 SAPICS 2015, www.sapics.org.za ISBN 978-0-620-64684-0 PAGE 5