Transcript Database Mining in Enterprise Resource Planning

Database Mining in Enterprise
Resource Planning
UCI Freshman Seminar
May 10th, 2004
Patrick Nguyen Huu
Introduction To ERP
• Enterprise Resource Planning is a software
architecture that facilitates the flow of
information among the different functions or
processes of an enterprise (manufacturing,
logistics, HR, supply-chain management, etc)
• ERP provides the backbone for an enterprisewide information system
• A database, in turn, is the backbone of an ERP
system
Client/Server Architecture
The ERP Software Industry
• Until recently, “JBOPS“ were market leaders (J.D.
Edwards, Baan, Oracle, PeopleSoft, SAP)
• “JBOPS“ -> “SOPMS“ (PeopleSoft acquired J.D.
Edwards, Oracel has a hostile bid for PeopleSoft, Baan>Invensys->SSA Global Tech)
• There is a number of specialized providers (Siebel
Systems for customer relationship management, i2 for
supply-chain mgmt)
• D-Base companies provide the backbone (Oracle,
MS(SQL), Sybase, IBM)
The Business Intelligence
Imperative
• Forces on the supply side:
– The Internet reduces barriers of entry in many markets
– Interactive customer channels and increased bandwidth
– Explosion of power and data storage capabilities
Forces on demand side:
Increasingly informed and empowered customers
Lower search costs
Higher customer acquisition costs
=> As a result, the leverage of corporate data resources is
becoming essential for business success
Tools For Responsive Marketing
• Clustering: Partitioning a dbase so that records within
each group are sufficiently
alike
• Profiling: Obtaining a
generalized classification
model from a database sample
• Factor Seeking: Finding
interdependencies in the data
based on associations
between groups of attributes
• Database Marketing:
Translates customer purchase
and demographic information
into an individual purchase
probability, by means of a
response model based on
purchase history, demographic
variables, and product
attributes
Companies Built Around D-Base
Marketing
• Calyx&Corolla – Marketer of floral arrangements: Analyzes customer
d-base to determine which customers to target with different flower
catalogs to balance prospecting costs against the lifetime value of
the customer
• Franklin Mint – direct marketer of quality collectibles and products:
Sends catalogs to only those customers with the highest propensity
to purchase the offered product; this increases the response rate,
lowers costs and increases profits
• Capital One – leading credit card issuer: Uses customer profitability
analysis to deliver the “right product to the right customer, at the
right time and at the right price“
How Database Marketing Works?
The Harrah‘s Case
• Harrah‘s Casino, Las Vegas:
– Difference form other casinos: focus on the middle-class,
medium-scale gamblers rather than the “high-rollers“
– Had several attempts at customerinitiatives, such as the “Gold
Rewards“ => Collected a bulk of customer data via these cards
(purchases, points of visits, etc)
– Realized that the bulk of their income came from small to
medium scale gamblers, utilized their database of customer data
to customize service for each individual cardholder
What Made Harrah‘s Success
• Harrah‘s is mainly a casino facility with slotmachines and
the occassional gambling tables. It lacks the glamorous
attractions and shopping centers of comparable casinos
in Las Vegas
• Unlike other casinos – who draw their main income
through the attractions and shops – Harrah‘s focuses on
creating its “loyal“ clientele and providing best service,
giving them incentive to return
• Service for their clientele is personalized, based on the
analysis of the data collected from the individual
cardholder