Day 7

1
Day 7

• Using Computer-based Applications
• to Redesign Business Processes

2
Learning objectives

• Context and overview for next phase
• Computer-based apps
• Their role in redesigned business processes
• A classification
• Infrastructure Components
• Architecture Relationships between components
• Major desirable features
• Focus TPS and DST (this week), BPMS (next week)
• Application exercise

3
Context
4
Planning process integration

• Activities
• Examine alternative IT (and IT) integration
  options
• Adjust process design
• Plan process implementation
• Deliverable Process integration plan
• Key participants IS design team BPR team
• Focus on the design of IT solutions that
• Support/shape critical BPs (such as procurement,
  production and sales)
• Allow for business integration (BPMS, EAI, ERP,
  etc.)

5
Overview of the role of IT

• A business process is a coordinated chain of
  activities consisting of
• Human activities (also known as human workflow)
• Business system (IT) activities (also known as
  automated activities)
• Combination of human and systems activities
• Map IT (and required data) to each task in the
  redesigned business process
• The role of IT in a redesigned business process
  is to
• Support/shape that business process
• Provide for the
• Execution of the redesigned process and the
  capture the process execution data
• Business activity monitoring (i.e. set of
  reporting tools for process owners/managers)
• User portal that allows end-users to participate
  in the process execution
• Admin portal used by process owners/managers to
  deploy and manage the process

6
Revisited context

• Done
• "Best" TO-BE process is selected
• Modeling of TO-BE defines IT requirements
• Next
• Use existing computer-based apps to support the
  TO-BE process
• AND/OR
• Define specs and build new computers based app(s)
  needed for the TO-BE process
• Then
• Implementation of the TO-BE process
• Maintenance of the IT application(s) and the
  TO-BE process
• ?Business Integration also known as BPM as the
  ultimate objective
• BPM Discipline of modeling, automating, managing
  and optimizing a business process (Khan , 2004)
• BPM implies the use of BPMS that execute the
  redesigned business process and monitor its
  performance

7
Computer-based apps a classification

• Classification based on hierarchy in
  organizations (Anthony 1965)
• Computer-based apps used at the operational
  level
• ?Operational systems or TPS Used to
  collect, store, process and disseminate data
  related to an organizations daily transactions
• ?Can be departmental systems (Ex
  ordering systems, inventory management systems),
  enterprise systems (Ex Workflow systems, ERPs)
  or inter-organizational systems (Ex EDI, VAN, AA
  Sabre, etc.)
• ? Typically developed using the SDLC
  methodology (software can be off-the-shelf)
• Computer-based apps used at the managerial level
  
• ?Managerial systems Used for short-term
  planning, organizing, control (Ex MIS, DSS)
• ? Typically developed using prototyping
• Computer-based apps used at the level of top
  managers Strategic-level systems Used to
  support decision made by top managers (?
  strategic systems) (Ex EIS)
• Other Knowledge Management System (KMS), Office
  Automation System (OAS), Groupware

8
TPS Basics

• Transactions
• Events that occur as part of doing business
• Sequence of database operations
• TPS (AKA OLTP)
• Include departmental/functional, cross functional
  or cross organizational systems
• Capture and process data arising from occurrence
  of business transactions
• Transaction processing cycle
• Data entry using POS terminals (optical scanners
  and card readers), clients (online ordering
  systems), etc.
• Transaction processing Real-time or batch
  processing
• Database maintenance Updates in databases follow
  occurrence of transactions
• Output, report generation and queries
• Database focused on operations optimized for
  updates ERD Normalization

9
IT infrastructure

• Hardware
• Microcomputers, servers and possibly mainframes
• Peripherals (bar code readers, scanners, digital
  cameras, printers, etc.)
• Software
• OS and NOS Typically Unix/Linux, Win
• Application software
• Other middleware, antivirus, etc.
• Network (ITEC 3210)
• LANs and BBNs switched-Ethernet as the best
  network architecture
• Leased MAN and WAN connections Several options
  SLAs
• Other Internet connections / Hosting services /
  VANs
• Database (ITEC 3220) Collection of logically
  related records

10
IT architecture

• Divides application layer functions among
  computers to deliver service to users
• Data storage Storage and retrieval of data
• Data access logic Query required to access data
• Application logic Work performed by application
• Presentation logic User interface
• ? Networked environment
• Often, the best practice is the client-server
  architecture
• Two-tier
• Three-3 tier
• N-tier Web systems
• Universal client with a browser
• Used for the Internet, intranets (corporate
  portals) and extranets
• In specific cases, the host-based architecture
  remains the best solution

Improved response time with an increased number
of users volume of transactions
11
Transactional web systems an illustration
(source www.dell.com)
12
Major TPS NF requirements

• Availability System must be up at all time
• Reliability System should rarely fail
  (fault-tolerant systems)
• Response timeYour turn
• Caching servers
• Load balancing servers
• Throughput Volume of transactions per second,
  peak
• Scalability Ability to expand without breaking
  down
• Security Your turn
• Other requirements
• Low cost Average cost per hour of operation
• Configurability for the above requirements

13
Analyzing data from TPS

• TPS generate large amounts of data stored in
  operational DB (DB for OLTP)
• Such data is used to guide the decision making
  process of managers
• Decision-support queries are more complex than
  transaction-oriented queries
• OLAP Compute the percent change in sales of
  pdct X for all of our stores in Ontario between
  1990 and 2000
• Cube A multidimensional way to analyze
  information (who, what, why, when, and where)
• Drill Down Going from a summarized view to a
  more detailed view of information within the same
  cube
• Approach
• Data warehouse (or data marts) Typical but NOT
  required as data can be mined from operational
  files
• DSS Model-driven (statistical models) or
  data-driven (detect unexpected relationships)

14
Data warehouse basics

• Databases for OLAP are different from OLTP
  systems
• Usually larger (TBs vs. GBs)
• Separated from OLTP systems to prevent
  interference
• Different users, hence different interfaces
• Different organization of data to support complex
  queries
• Characteristics of data warehousing
• Data are organized by subject Customer, vendor,
  product, price level, etc.
• Data are kept for many years
• Data are consistent (same meaning for codes over
  the years and the tables)
• Data warehousing process
• Modeling data for the data warehouse
  Denormalization Renormalization (Star)
• Extracting data from operational databases
• Cleaning and summarizing extracted data
• Loading the data into the data warehouse database
  (typically include external data)

15
The star structure

• The most common dimensional model
• A fact table surrounded by dimension tables
• Fact table contains FK for each dimension table
• Dimension tables reference data
• All tables are read-only (when accessed by users)

16
Using a data warehouse

• Report generation
• Trend detection, fraud alerting,
• Interactive queries
• Find facts that are in the data
• OLAP (OnLine Analytical Processing)
• Focused examination of data
• "Drill down" to get more detail
• Visualize relationships between data (statistical
  models)
• Data mining
• Discovery of knowledge hidden in the data
• Prediction of unexpected relationships

17
Application exercise Your project

• Indicate for each computer-based app. required
• - Its type (see typical classifications)
• - Its functional requirements
• Translate your business needs into IT design
  specs
• - Architecture
• - Infrastructure
• ERD Just focus on
• - Tables (DB) needed
• - Reports needed

18
Thank you