Managing GIS

1
Managing GIS

• Longley Ch. 17

2
Information Sources

• Tomlinson, Roger Thinking about GIS GIS
  Planning for Managers ESRI Press, 2003
• Zeiler, M. Modeling our World The ESRI Guide to
  Geodatabase Design ESRI Press, 1999
• Huxhold, William E. and Levinsohn, Allan G.
  Managing Geographic Information System Projects
  New York Oxford, 1995
• Harmon and Anderson The Design and implementation
  of Geographic Information Systems Wiley, 2003
• Obermeyer, Nancy J. and Pinto, Jeffrey K.
  Managing Geographic Information Systems New York
  The Guilford Press, 1994 (dated and very
  academic)
• von Meyer, Nancy and R. Scott Oppman Enterprise
  GIS.URISA, 1999, 98 pp. (set of case studies)
• Derek Reeve, GIS, Organizations and People
  London Taylor Francis, 2000 (UK case studies)

3
The problem

• Graduate student wants to use spatial analysis on
  field data for health study
• Small consulting company with massive collection
  of site plans wants to automate delivery of plans
  to developers to save money
• Government of country wants to support 2010
  census with automated mapping, analysis and
  Internet-based publishing/searching of new data

4
Which level of GIS?

• Project
• Single department application (Departmental GIS)
• Multi departmental application
• Enterprise system (Enterprise GIS)
• Multi Organizational endeavor (Community GIS)

5
Example Enterprise

• Debra Gondeck-Becker, Assoc. AIA Jordani
  Consulting Group
• Implementing an Enterprise-Wide Space Management
  System - A Case Study at the University of
  Minnesota
• 1999 Study with Test Implementation

6
Conceptual Design
7
Software Selections
8
Building/Room level management
9
Power, facilities etc.
10
Level I Project

• Organizational Environment
• Expected result is a product, e.g a map or report
• Project has an end date an finite project
• No long-term support expected no commitment to
  ongoing GIS
• Little organizational impact
• GIS Implementation Approach
• One-time effort
• Need best tool for the job
• Consultant or contractor may do entire thing

11
Level II Single Department

• Organizational Environment
• Small Organization or Single Department
• Well-defined, existing business function to be
  supported
• Ongoing support is required but no commitment to
  GIS
• Little or no reorganization e.g. manual drafters
  shift to GIS workstation
• Managed by departmental responsible for business
  activity
• GIS Implementation Approach
• PC or standalone workstation
• maybe CAD focused
• Little or no integration with attribute databases
• Little sharing of information within or beyond
  department

12
Level III Multi-Department

• Organizational Environment
• Mid-size to large organization, more than one
  department
• More significant commitment of staff and budget
  to GIS
• Ongoing support and update strategies
• Some organizational implications (Champion)
• Managed by cooperating departments
• GIS Implementation
• Multiple, networked PCs/workstations
• Topological GIS
• Object/Relational database
• Some information sharing between departments

13
Level IV Enterprise System

• Organizational Environment
• Usually medium to large organization, multiple
  departments
• High level long-term commitment to GIS
• Organization-level strategic planning,
  distributed implementation and maintenance
• Incorporation of GIS as part of organizational
  infrastructure
• Corporate management support and involvement is
  essential
• GIS Implementation
• Distributed client-server network(s)
• Integration of multiple GIS, database, and
  related technologies
• Multi-department data sharing, standards and
  metadata

14
Level V Multi-Organizational

• Organizational Environment
• Public organizations or industry alliance (See
  Ch. 20 on Partnerships)
• Multi-participant organizational structure for
  planning and policy
• Distributed maintenance responsibilities across
  organizations
• Long-term, high level commitments among
  participating organizations
• Significant reorganization of functions across
  organizations
• GIS Implementation
• Distributed maintenance of shared elements
• Data exchange facility and standards and
  metadata, Internet or other WAN
• Data integration from multiple technologies

15
Management Responsibilities

• Planning
• Choice
• Design
• Strategy
• Staffing
• Standardize
• Interoperability
• Reuse
• Document
• Assume personnel loss and turnover
• Write formal documents

16
Ten step GIS Planning MethodologyTomlinson,
Thinking About GIS

• Consider the strategic purpose
• Plan for the planning
• Conduct a technology seminar
• Describe the information products
• Define the system scope
• Create a data design
• Choose a logical model
• Determine system requirements
• Benefit-cost, migration and risk analysis
• Make an implementation plan

17
Analysis of Requirements
Evaluation of Alternatives
Specification of Requirements
Implementation of System
6. Final Design
7. Request for Proposal (RFP)
14. Implementation
A Fourteen Step Implementation Process!
(assumes external acquisition)
Source Longley, et. al. p. 391
18
Gantt Chart
19
GIS Implementation issues

• GIS Paradigm
• Use of spatial location as integrating framework
  for information
• Power of spatial analysis
• Geographic Data Management Principles
• Extend data management principles to include
  geography
• Builds on standard IT practice
• Technology
• Select appropriate GIS-enabling technology and
  plan to evolve
• Follow and exploit new technologies
• Organizational Setting
• Organizational setting a crucial ingredient to
  success/failure
• Level and nature of enterprise

20
Pilot Projects

• As a demonstration, to show potential users the
  possible utility of GIS
• As an experiment to test a particular technical
  aspect of implementation
• As a temporary operation or production
  environment to assess operational feasibility or
  to determine organizational impact
• As a trail run to test adequacy of project
  planing and design
• As a benchmark test to compare hardware,
  software, network configurations being considered

21
Resources for Developing a GIS

• Developing a GIS involves investment in five
  areas computer hardware, computer software,
  geographic data, procedures and trained staff.
• Developing the geographic database (which
  includes some of the procedure and staff costs)
  can account for 60 to 80 of the GIS development
  cost.

22
Staffing Requirements for a GIS

• Three areas where expertise is needed includes
  management of the GIS project (GIS project
  manager), GIS database skill (database
  administrator), and application development for
  database and users (GIS software analyst).
• In the case that the three experts are hired, a
  full-time GIS manager is available on staff.
• Alternatives to staff expansion are consultants
  and data conversion firms.

23
System development team
Seconded from line business unit?
Seconded from or hired by IT?
Full time!
Technical staff IS GIS programmers, database,
computer operators, cartographers, data entry
Support Staff administrative, secretarial
24
System Development TeamMedium or small sites
Business Expertise
25
GIS Development Cycle
First decides what the GIS should do, second
decide how the GIS will accomplish each task.
Needs Assessment
Conceptual Design
Database Planning and Design
Database Construction
Available Data Survey
GIS Use and Database Maintenance
GIS System Integration
Application Development
Pilot / Benchmark
Acquisition of GIS HW and SW
HW and SW Survey
26
Process Flow

• Needs assessment
• System conceptual design
• Survey existing data
• Select and implement hardware and software
• Implement
• Evaluate and improve

27
Needs Assessment

• Data/map inventory is not always a wise approach
  (legacy approach)
• Interviews, focus groups can capture the needs of
  a dept (managers, users, customers)
• Compiling the results of the needs assessment
• Master data list
• Master function list
• Budget constraints
• Assess available systems
• Create matrix
• Score systems (yes, no or ratings)
• Select system with highest score

28
Example Paper walkthrough
GIS Application Description Form (New York State
Archives)
Application Identification
Description
Functions
Plateau or Terrace
Entities Attributes
Lower Slope
Rise
29
Conceptual Design

• Data needs identification from the needs
  assessment
• Selection and creation of the data model
  (including metadata)
• Workflow plan for entering data into the
  database
• Updating and maintenance

30
Conceptual Design
Life Cycle of a GIS Database Source Documents
Source Documents Maps, Images, Air Photos, etc.
Data Objects Identified During Needs Assessment
Preparation of Data Model
Match Needed Data to Available Data and Sources
Survey and Evaluation of Available Data
Prepared Detailed Database Plan
Map and Tabular Data Conversion
Create Initial Metadata
Add Record Retention Schedules to Metadata
Database QA/QC Editing
Shelf
GIS Database
Lower Slope
Continuing GIS Database Maintenance
Database Backups
Archives
(courtesy of New York State Archives)
31
Describe Formal Design
Simple Entity - Relationship (E-R)
Diagram Entities represented as rectangles,
relationship as diamonds and attributes as
ellipses.
Building Located on
Parcel Resides Owned
by Occupant John Smith Owner
Acme Corp.
Shelf
Upper Slope
Plateau or Terrace
Deep Ocean
Lower Slope
Rise
32
Need for Metadata

• Information about data
• describe the characteristics of the data (entity
  and attributes)
• provide information on accuracy and source
• Functions of metadata
• a basic data description of a data set
• information for data transfer/sharing
• information for entries into clearinghouses to
  catalog availability

33
FGDC Metadata standard
34
Example FGDC Reference Section
35
Survey of Available Data

• Inventory maps, tables, digital or analog
• Consider all useful (air photos, remote sensing,
  DEMs, vector maps, CAD files, scanned images,
  survey data, field data, statistical data
• Document availability Source, costs,
  redistribution rights, licenses, etc.
• Document accuracy, precision, date, consistency
• Identify, acquire, reformat, transform
• Ingest

36
GIS Hardware and Software Selection

• Status of the current hardware and software
  market
• What are other people (consulting firms,
  universities, governments) using?
• Note hardware and software combinations
• Local data formats and data conversion
  capabilities

37
Survey of GIS Hardware and Software

• Select Software first
• Evaluate software functionality and performance
• Conduct tests and benchmarks
• Test systems integration Use real data
• Select hardware to suit software, plan on 2-3
  year HW cycle other than special devices
• Memory and disk are cheap
• Make choice and purchase
• Consider licensing, maintenance, training etc.

38
Large Scale Hardware and Software Procurement

• Identify an Evaluation Team
• Research purchasing rules processes
• Obtain samples of RFPs
• Define evaluation criteria include in RFP
• Prepare Request for Proposal (RFP)
• Distribute RFP to potential vendors
• Hold Bidders Meeting
• Evaluate Proposals
• Select Winner and Write Contract

39
GIS Development GuidesState of New York, Local
Government Technology Services (1997)
http//www.sara.nysed.gov/pubs/gis/gisindex.htm
1
2
6
5
3
11
9
9
8
7
4
40
Issues During ImplementationCampbell, (1992)

• Technological, associated with system
  compatibility
• Data-related, associated with lack of consistency
  between data sets
• Organizational, associated with data ownership
  and control
• Institutional, associated with how to use data in
  the policy-making process

Time
41
Institutional FactorsCampbell, 1992

• Organizations, and units in them, jealously guard
  their scope of activity and treat with suspicion
  proposals that may change this
• Applications that give cost savings are more
  readily accepted than decision-making
  applications
• Local communities very suspicious of developments
  that suggest centralization of information and
  therefore power
• GIS techies often uncomfortable with social and
  political aspects of system implementation and
  utilization, thus need to involve
  politically-adept users/line managers/policy
  makers

42
and the process is endlesssource Longley, et.
al. p. 390