INFS Chapter 8-11

Organizations can find their most valuable customers through...

"RFM" - Recency, Frequency, and Monetary value

Source (scm)

-Companies must carefully choose reliable suppliers that will deliver goods and services required for making products -companies must also develop a set of pricing, delivery, and payment processes with suppliers and create metrics for monitoring and improving the relationships

Dependent Data Mart

-Doesn't have its own source systems -data comes from the data warehouse -dependent data marts provide users with a subset of the data from the data warehouse, in cases when users or applications don't want, need, or are not allowed to have access to all the data in the entire data warehouse

ETL Infrastructure

-Infrastructure that facilitates the retrieval of data from operational databases into the data warehouses -Includes the following tasks: -extracting analytically useful data from the operational data sources -transforming such data so that it conforms to the structure of the target data warehoure model, while ensuring the quality of the transformed data

OLAP/BI Tools

-OLAP tools are designed for analysis of data in data warehouses and data marts -OLAP tools are also known as Business Intelligence (BI) tools, and are often referred to as OLAP/BI tools -they allow users to retrieve needed date from data warehouses and data marts by using simple point-and-click mechanisms -based on the point-and-click actions by the user of the OLAP/Bi tool, the tool writes and executes the code in the language of the DBMS that hosts the data warehouse or data mart that is being queried

Accessing Data in Data Warehouses and Data Marts

-OLAP/BI tools -front-end/BI Applications -executive dashboards

Source Systems

-Operational Databases that provide analytically useful information for the data warehouse's subjects of analysis -2 Purposes: original operational purpose, as a source system for the data warehouse -can include external data sources EX: market research data, census data, stock market data, weather data

RFID in Retail Supply Chain

-RFID tags can be added to every product and shipping box -at various stops of an item's journey (e.g. manufacturer, distribution center, store) a reader scans tags and updates information in the computerized SCM system

OLAP/BI Tools

-Slice and Dice -Pivot (Rotate_ -Drill Down / Drill Up

Data Warehouse

-Sometimes referred to as the target system b/c it is the destination for the data from the source systems -a typical data warehouse periodically retrieves selected analytically useful data from the operational data sources

Data Warehouses

-Structured and modeled differently than operational databases -data warehouses use different types of technologies for storage and retrieval than operational databases -data warehouses are used by different (much smaller) set of users that operational databases: -operational database can be used by thousands of employees, partners and customers -while, data warehouses are used by data analysts and decision makers

Visibility (fdscm)

-ability to view all areas up and down the supply chain -not possible without IT -bullwhip effects: occurs when distorted product demand information passes from one entity to the next throughout the supply chain (can result in excessive or insufficient inventory)

Speed (fdscm)

-as the pace of business increases, and organization's supply chain must respond efficiently, accurately, and quickly (both the increases in the pace of business, and the consequent responses in the supply chain are IT driven)

Data Mining Pt 2

-data mining is an intersection of database management, artificial intelligence (machine learning) and statistics -data warehouse is one of the primary sources for data mining

Competition (fdscm)

-increased competition and adoption of IT advances in supply chain management, makes any organization that is ignoring its supply chain modernization at risk of becoming obsolete

CRM entails Integration of:

-management of sales -marketing -customer support

Data Warehouse Front-End Applications

-often referred to as BI (Business Intelligence) applications -used to provide access to the data warehouses for users who are engaging in indirect use

OLAP / BI Tools Functionalities

-slice/dice, pivot, drill -Data Visualization: graphically visualizing the answers -Additional Functionalities: creating and examining calculated data, determining comparative or relative references, performing exception analysis, trend analysis, forecasting, and regression analysis -number of other analytical functions

Data Warehouse Components

-source systems (operational databases) -extraction-transformation-load (ETL) infrastructure -data warehouse -front-end (BI) applications

Independent Data Mart

-stand alone data mart, created in the same fashion as the data warehouse -Independent data mart has its own source systems and ETL infrastructure

deliver (scm)

-step is referred to as logistics -logistics is the set of processes that plans for and controls the efficient and effective transportation and storage of supplies from suppliers to customers -during this step, companies must be able to receive orders from customers, fulfill the orders via a network of warehouses, pick transportation companies to deliver the products, and implement a billing and invoicing system to facilitate payments

PLAN (scm)

-strategic portion of supply chain management -A company must have a plan for managing all the resources that go toward meeting customers demand for products or services -a big piece of planning is developing a set of metrics to monitor the supply chain so that it is efficient, costs less, and delivers high quality value to customers

make (scm)

-this is the step where companies manufacture their products or services -this includes scheduling the activites necessary for production, testing, packaging, and preparing for delivery -this is by far the most metric-intensive portion of the supply chain, measuring quality levels, production output, and worker productivity

Return (scm)

-this is typically the most problematic step in the supply chain -companies must create a network for receiving defective and excess products and support customers who have problems with delivered products

Supply Chain's 3 Main Links

1. Materials flow from suppliers and their "upstream" suppliers at all levels 2. Transformation of materials into semi-finished and finished products through the organization's own production process 3. Distribution of products to customers and their "downstream" customers at all levels

Reasons for Creation of a Data Warehouse As a Separate Analytical Database:

1. Time Horizon Difference 2. Performance Reasons 3. Difference in Structuring of Data

Data Mart

A data store based on the same principles as a data warehouse, but with a more limited scope

Reader (RFID)

A reader uses radio waves to read the tag and send the EPC to computers in the supply chain

Data Warehouse

A separate data store with primary purpose of data analysis.

Slice and Dice

Adds, replaces, or eliminates specified attributes (or particular values) from the already displayed result

Market Basket Analysis

Association rule: X to Y -Correlations are expressed in the form: transactions that contain X are likely to contain Y are well -EXAMPLE: Customers buying X are likely to buy Y during the same purchase transaction

Customer Relationship Management

CRM is a business philosophy based on the premise that those organizations that understand the needs of individual customers are best positioned to achieve sustainable competitive advantages in the future

Analytical Database (Data Warehouse)

Collects and presents analytical information in support of analytical tasks

Operational Database

Collects and presents operational information in support of daily operational procedures and processes. Typical for organizations to have many.

1. Time Horizon Difference

Difference between how long the data is useful for transactional vs. Analytical needs. -regular transaction oriented databases store data for a limited time (60-90 days) before the data loses its immediate usefulness and it is archived -keeping data in transaction-oriented databases for longer periods of time would simply cause database operations to deal with unnecessarily large amounts of old not longer used data -on the other hand, data warehouses tend to keep years worth of data in order to enable long-term analysis

Computer Network (RFID)

Each computer in the supply chain recognizes the EPC and pulls up information related to the item such as dates made and shipped, price, and directions for use, from a server maintained by the manufacturer. The computers track the item's location throughout the supply chain.

Market Basket Analysis (a.k.a.Association Rule Mining)

Finds group of items that tend to appear together in transactions

Monetary Value

How much a customer spend on each purchase

Analytical Information

Information collected and used in support of analytical tasks. This is based on Operational (transactional) information -EX: information reflecting trends, sales, product statistics, future growth projections

3. Difference in Structuring of Data

It is often impossible to structure a database which can be used in an efficient manner for both operational and analytical purposes

Drill Up

Makes the granularity of the data in the query coarser

Drill Down

Makes the granularity of the data in the query finer

Market Basket Analysis Part 2

Measures: -Support= (transactions containing X and Y) / (# of all transactions) -Confidence (number of transactions containing X and Y) / (# of transactions containing X)

2. Performance Reasons

Performance of operatuional day-to-day tasks involving data use can be severely diminished if such tasks have to compete for computing resources with analytical tasks

Manufacturer

Reader scans the tags as items leave the factory

Distribution Center

Readers in the unloading area scan the tags on arriving boxes and update inventory, avoiding the need to open packages

RFID System Components

Tag, Reader, Computer Network/System

Store

Tags are scanned upon arrival to update inventory. At the checkout counter, a cashier can scan individual items with a handheld reader. As items leave the store, inventory is updated. Manufacturers and retailers can observe sale patterns in real time and make swift decisions about production, ordering, and pricing.

Supply Chain Planning (SCP) Software

Uses advanced mathematical algorithms to improve the flow and efficiency of the supply chain

Radio Frequency Identification (RFID)

Uses electronic tags and labels to identify objects wirelessly over short distances

Factors Driving Supply Chain Management (fdscm)

Visibility Competition Consumer Behavior Speed

Tag (RFID)

a microchip holds data, in this case an EPC (electronic product code), a set of numbers unique to an item. The rest of the tag is an antenna that transmits data to a reader. EX: 01-0000A77-00136BR5

Supply Chain Execution (SCE) software

automates the different steps and stages of the supply chain

Consumer Behavior (fdscm)

companies can respond to demanding and changing customers through supply chain enhancements enabled by IT (e.g. demand planning software using statistical and forecasting techniques)

Data Mining

discovering novel and interesting patterns in large amounts of data -Patterns should be: accurate, meaningful, understandable, actionable

Using Executive Dashboards

for a quick glance at data -Intended for use by higher level decision makers within an organization -contains an organized easy-to-read display of a number of critically important queries describing the performance of the organization -in general, the usage of executive dashboards should require little or no effort or training -executive dashboards can be web-based

Using OLAP/BI Tools

for direct access

Using Front-end/BI Applications

for indirect access -front end applications accessing analytical data are also know as BI applications -most users who need to access analytical data in data warehouses and data marts can not engage in direct access (as they don't know how to use OLAP/BI tools) Instead they are given access to front end/BI applications

Frequency

how frequently a customer purchased items

Recency

how recently a customer purchased items

Operational Information (transactional)

information collected and used in support of day to day operational needs in businesses and other organizations. (EX: withdrawing cash from an ATM, making an airline reservation, purchasing stocks)

Data Mining techniques

many techniques -one widely used one is Market Basket Analysis

Confidence

measure the strength of the correlation, so rules with low confidence are not meaningful, even if their support is high

Support

measures the significance of the rule, so we are interested in rules with relatively high support

Online Analytical Processing (OLAP)

querying and presenting data from data warehouses and/or data marts for analytical purposes

Pivot (Rotate)

reorganizes the values displayed in the original query result by moving the values of a dimension column from one axis to another

analytical CRM

supports back-office operations and strategic analysis and includes all systems that do not deal directly with the customers

Operational CRM

supports traditional transactional processing for day-to-day front-office operations or systems that deal directly with the customers

Online Transaction Processing (OLTP)

updating (i.e. inserting, modifying and deleting), retrieving and presenting data from database for operational purposes

Predictive Analytics

using past data to predict future events