Docstoc

Project Management as a Breakthrough at CERN

Document Sample
Project Management as a Breakthrough at CERN Powered By Docstoc
					     Achieving New Goals and Results
         Using a Project Approach

            B.Denis, M.Vanden Eynden
                  CERN - SL/CO
               Geneva, Switzerland



28/11/97      B.Denis, M.Vanden Eynden - CERN   1
                         SL Seminar
                As Introduction ...
   “As far as possible the scientific program should be divided
    into projects with finite lifetimes and defined objectives or
    end points. These may include some long-term strategic
    projects and can be of a wide variety of sizes. There should be
    clear procedures for establishing, reviewing and terminating
    projects.”


                                               Geneva, December 1987




    28/11/97            B.Denis, M.Vanden Eynden - CERN                2
                                   SL Seminar
                       Outline
   Facing a Challenging Future
   The Current Working Practices
   Considering New Ways of Organizing Activities
   Concepts of Project Management
   Applicability to Contracted Developments
   Benefits
   Drawbacks
   Proposal




    28/11/97          B.Denis, M.Vanden Eynden - CERN   3
                                 SL Seminar
             Facing a Challenging Future
   More performance, more technology
    – but also more complexity
         nowadays controls problems are rarely of such nature that they
          can be resolved at one specific place in the base organization.
         Effective collaboration and clear responsibilities are essential

   Tight Budget Conditions, less human resources
     – but big projects to come (LHC, …)
           working practices should be improved and adapted to this new
            reality
   More activities will be contracted out to industry
    – but who is ready to work this way ?
           CERN engineers should understand and develop a culture of
            projects already widely spread in industry

      28/11/97               B.Denis, M.Vanden Eynden - CERN               4
                                        SL Seminar
               Current Project Practices

   What is a project ?
    ‘A project
    – is a unique task (risky)
    – is designed to attain a specific result
    – requires a variety of resources (new organization)
    – is limited in time’


   What kind of project are we talking about ?
    – small and medium-size projects
    – activities having the characteristic features of a project.


    28/11/97               B.Denis, M.Vanden Eynden - CERN          5
                                      SL Seminar
     Current Project Practices (Cont.)
   Working groups
    – absence of formal organization


   Regular meetings
     Minutes used to plan the work between meetings


   Absence of formalism
     fear of bureaucracy


   Tools (if any) used as scheduling aids
     MS project often used to produce schedules that are never updated

    28/11/97             B.Denis, M.Vanden Eynden - CERN           6
                                    SL Seminar
Considering New Ways of Organizing
             Activities
   1993
    – SSIS project (SL/CO + SL/OP - closed in January 1995):
              development of a new software interlocks system for the SPS
              contracted out to industry
                  use of ESA PSS-05
                        structured specifications
                        project management practices imposed to the contractor
                        formal project reviews and acceptance tests
   1994
    – PowerPC project (SL/CO - closed end 1996)
              re-engineering of the LEP front-end infrastructure (HW+SW)
              in-house
                   ESA-like Project Management Plan
                   formal project reviews
    28/11/97                     B.Denis, M.Vanden Eynden - CERN                  7
                                            SL Seminar
Considering New Ways of Organizing
          Activities (Cont.)

   1996
     – CIS project (SL/PC + SL/CO - open)
               development of a new control and interlock system (HW and SW) for
                the SPS main power converters
               contracted out to industry
                   structured, ESA-like specification


   Need for a more pragmatic approach. Need to define more complex
    and realistic lifecycles without forgetting the benefits of ESA PSS-05
    (especially for contractual situations).



     28/11/97                   B.Denis, M.Vanden Eynden - CERN                8
                                           SL Seminar
Considering New Ways of Organizing
          Activities (Cont.)
   1997
    – SPS 2001 project (SL/CO + SL/OP - open)
              development of a new application suite for the SPS in the LHC era
              in-house (?)
                   first GDPM project definition workshop
    – Apollo97 project (SL/CO + SL/OP - closed yesterday)
              eradication of Apollo workstations
              collaboration with IHEP Protvino
                  full adoption of GDPM approach
    – SCAM project (SL/CO + ST/MC - open)
              development of a software configuration system
              in-house project including the selection of a commercial tool
                   project definition workshop
                   GDPM approach
                   tool specified using ESA User Requirement Document
    28/11/97                    B.Denis, M.Vanden Eynden - CERN                    9
                                           SL Seminar
         Concepts of Project Management
    Organizing activities as projects :
                                                     Defining objectives &
     – Five functions :
                                                       building strategy

                                       Plan


                Control                Lead                   Organize


                                  Implement                      Defining tools,
  Tracking &                                                     responsibilities
problem solving              Identifying &                       and techniques
                          performing activities

     28/11/97               B.Denis, M.Vanden Eynden - CERN                  10
                                       SL Seminar
 Concepts of Project Management (Cont.)
    Project Planning and Organization                              Project Steering
  Project                                                            Committee or
DefinitionW                   Define Project                         Memorandum
 orkshop                 (Purpose, Scope, Objectives)


      Build Project           Define Managerial                Define Technical
       Strategy                    Process                         Process
     Project Model &         Monitoring Mechanisms                  Tools

    Milestone planning            Staff Planning                  Techniques

                               Procedures (reviews)             Documentation


                                Define Interfaces
                                        &
                                Responsibilities
      28/11/97               B.Denis, M.Vanden Eynden - CERN                    11
                                        SL Seminar
Concepts of Project Management (Cont.)
   Project Planning and Organization - Define Project

    – Pitfalls
              People affected by the project don’t agree on the goals and
               scope of the project
              People involved in a project are not full time available
              Planning is too detailed and never updated
              Wrong information to the wrong persons
    – Project Definition Workshops
              Aim : build a common understanding and commitment to the
               project objectives and responsibilities
              Important People : end users, project members, line managers
               and people responsible for the operation of the future
               system(s)
    28/11/97                  B.Denis, M.Vanden Eynden - CERN                12
                                         SL Seminar
 Concepts of Project Management (Cont.)
   Project Planning and Organization - Build Project Strategy

Project Model       Milestone Plan              Responsibility Chart
                     M       T       O                 Execute the work (X)
                                                       Take Decision (D,d)
     Phase 1                                           Manage Progress (P)
                                                       Provide Tuition (T)
                                                       Must be consulted (C)
      Phase 2                                          Must be informed (I)


        Phase 3                                       Activity Schedule
                                                     Activity 1
          Phase 4
                                                                  Activity 2

    GDPM Approach                                                              Activity 3

     28/11/97            B.Denis, M.Vanden Eynden - CERN                          13
                                    SL Seminar
         Concepts of Project Management (Cont.)
         Project Planning and Organization - Milestone Planning

                                                          MILESTONE PLAN - pre-development
               Organisat ion   Management
                                            Technical
                                            D sions
                                             eci
                                                        Project:Apollo97
               and              eci
                               D sions
               Document at i
               on

Planned date        O             M               T     Milestone
2/19/97                                                 Projec t objec tiv es , s c ope and res pons ibilities are approv ed
                                  M1

3/9/97                                                  A lis t of Apollo s oftw are modules ex is ts and is v alidated by all SL groups
                   O1

3/23/97                                                 A tec hnic al propos al for the replac ement of the Apollo C -Tree databas e ex is ts
                                                 T1

3/16/97                                                 For eac h Apollo s oftw are, a referenc e to the doc umentation ex is ts
                   O2

3/3/97                                                  A tec hnic al propos al for the replac ement of the Apollo Graphic s s y s tem ex is ts
                                                 T2

3/23/97                                                 The terms of a pos s ible c ollaboration w ith IH EP Protv ino ex is ts
                                  M2

3/31/97                                                 D ec is ion to embark on the migration phas e is tak en
                                  M3




          28/11/97                                               B.Denis, M.Vanden Eynden - CERN                                           14
                                                                            SL Seminar
      Concepts of Project Management (Cont.)

        Project Implementation - Activity scheduling


                          ACTIVITY SCHEDULE - Dialog and Dataviewer
SCHEDULE                                                                                                  RESPONSIBILITIES
Project:Apollo97                           X   - executes the work




                                                                                                                                Eynden
                                           D   - takes decision solely
Milestone plan:pre-development             d   - takes decision jointly
No:T2




                                                                                                                          orpurgo
                                           P   - manages progress




                                                                                                                         s Robin

                                                                                                                         anden
 Weeks              Feb - 1997             T   - provides tuition on the job
                                           C   - must be consulted




                                                                                                                  Gulio M
                                                                                                                  George

                                                                                                                  Marc V
                                           I   - must be informed
                                           A   - available to advise
Estimate18 19 20 21 22 23 24 25 26 27 28 1 Code Activities                                                         1 2 3
                                                   Identify graphics tools in Apollo SW                            X X
                                                   Identify technical migration sc enario for each tool            X X C
                                                   Prepare Technical proposal                                      X X




          28/11/97                                B.Denis, M.Vanden Eynden - CERN                                   15
                                                             SL Seminar
           Applicability to Contracted
                 Developments
          Contracting the development of a system to industry is first of
           all a CERN project

 CERN’s project                Project leader
                                                              Buyer
                                                              (SPL)
                                         Other CERN staff
                   Core team             involved



 Contractor’s project
                               Project leader
                                                             Contract manager
      Quality manager

                                   Development team



28/11/97                   B.Denis, M.Vanden Eynden - CERN                      16
                                      SL Seminar
                    Applicability to Contracted
                      Developments (Cont.)
                   Milestone plan : allow the contractor to use its expertise,
                    specify ‘what’ and not ‘how’
Technical     Procurement    Management and
                               validation     Milestones

                                 M1           Project is defined

    T1                                        System specifications documented and approved
                    P1                        Market survey done
                    P2                        Contract awarded
                                 M2           System design approved
    T2
                                              CERN’s infrastructure ready for integration
                                 M3           System commissioned
                                 M4           System finally accepted

         28/11/97                     B.Denis, M.Vanden Eynden - CERN                         17
                                                 SL Seminar
                 Applicability to Contracted
                   Developments (Cont.)
                ‘Responsibility chart’ is a good tool to clearly specify contractual
                 relationships
                             X - executes the work
                                                    CERN             Contractor
                                 D - takes decision
                                 P - manage progress
                                 C - must be consulted
                                 I - must be informed
                                 A - available to advise
        Milestone
        System is commissioned                             DX   X     XC   PX   X

        Activities
        System tests                                       I               P    X

Provide the cabling to connect to CERN network             P    X          I

        Integration test                                        XA   C     PX   X
        Graphical interface validation                                XA   PX   X

        Acceptance tests                                   DX   X     X    PX   X

      28/11/97                           B.Denis, M.Vanden Eynden - CERN            18
                                                    SL Seminar
                         Benefits
                                                     Better communication
Organization                                         True commitment
                                                     Real transparency

                                                     Better visibility on work
       Line Managers                                 progress
                                                     Better control over costs
                                                     and resources


               Engineers                                         ?

                                                     Safer approach for new
                                                     technology insertion
                          Technical                  A good way not to rush
                                                     prematurely in endless
                                                     developments
  28/11/97         B.Denis, M.Vanden Eynden - CERN                        19
                              SL Seminar
                        Benefits (Cont.)
   Project Definition Workshops (13 People)
    – Level of satisfaction rated from 1 to 6 :

              The utility of the workshop : 4.75
              GDPM method : 4.6

   Strongest Points of GDPM Method and Workshops

              It promotes the definition of clear objectives : 12/13
              Planning as a group activity : 6/13
              One page report : 10/13



    28/11/97                   B.Denis, M.Vanden Eynden - CERN          20
                                          SL Seminar
                                 Benefits (Cont.)
Sample of 14 engineers distributed in 3 controls projects
                                           FALSE                                      TRUE

   It encourages good objectives definition                                                 94


 It encourages good definition of work and
                                                                                       89
              responsibilities


           It encourages good specification                                       83



         I encourages good project reviews                                        82



 It encourages clear definition of interfaces                                    79



          It builds a team spirit and I like it                             74



    I contributed to large comon objectives                            66


       28/11/97                      B.Denis, M.Vanden Eynden - CERN              21
                                                SL Seminar
                                Benefits (Cont.)
Sample of 14 engineers distributed in 3 controls projects
                                        FALSE                           TRUE
      It generates too much bureaucracy                            34

  It is only a layer on top of our practices                      31

    My work was not visible from outside                     24

We always did it differently and it worked                   24

                   It restricts my creativity           14

               It restricts my independence             13

                     I prefer to work alone         9

               I was not the project leader     4


    28/11/97                     B.Denis, M.Vanden Eynden - CERN         22
                                            SL Seminar
                            Benefits (Cont.)
Sample of 14 engineers distributed in 3 controls projects
 Would you like
 to see this new                                              86
   approach of
  project-based
  management
being developed        14
        ?                                                                 YES
                                                                          NO
In you have the
                                                                          YES BUT …
  choice in the                                               86
 future, would
 you prefer to
  work in this
 project-based         14
    context ?                                                            % of people
                   0   20         40         60         80         100


    28/11/97                B.Denis, M.Vanden Eynden - CERN                      23
                                       SL Seminar
                       Drawbacks
   Loss of creativity ?

    – Maybe but:
        Creativity is not required for every activity

        The level of formalism should be adapted to the nature of the

         project (methods must support the project and not constrain it)

    – And:
        milestone planning is a group activity and creative process

        planning by milestones allows the developers to use their

         creativity to reach their milestones



    28/11/97               B.Denis, M.Vanden Eynden - CERN             24
                                      SL Seminar
                 Drawbacks (cont.)
   Development time increase

    ‘We do not seem to have the time to plan the work adequately, but
      we seem to have time to do it twice’

    – In some case yes:
         But it is the price to pay for commitment, communication,

          transparency and knowledge transfer
         We are still in a learning phase and most people are not familiar

          with the approach

    – Not if you compare the time to produce a stable product


    28/11/97              B.Denis, M.Vanden Eynden - CERN             25
                                     SL Seminar
                  Drawbacks (cont.)
   Loosing the benefits of poor communication ?
       - Minimizes impact of poor planning (don’t let others know you do not
       know what you are doing)
       - Cuts down on questions, permits faster decision making, minimizes
       objections
       - Easier to deny what you said later on; preserves freedom to change you
       mind
       - Often a technique for gaining and/or maintaining power
       - Good technique to mask true intent
       - Helps you preserve mystique and hide insecurities
       - Allows you to say no nicely
       - Helps you avoid confrontation and anxiety
       - Avoids the need to share credit for your ideas
       - Encourages creativity; too much communication hampers thinking
       - Helps minimize opposition and criticism
                                                           (Bob Graham, 1989)

     28/11/97              B.Denis, M.Vanden Eynden - CERN              26
                                      SL Seminar
                         Proposal
   Introduce project-based management
    – That is : organizing activities as projects
    – Activities such as :
        New technical developments

        Contracts with industry

        Re-engineering work

        Non-technical aspects like for example the improvement of
          services
   Pre-conditions
    – “A sign from the Management” …
    – Understanding the “essence” of project management (objectives,
      planning, control) - a new way of thinking and working
    – High level of participation


    28/11/97            B.Denis, M.Vanden Eynden - CERN              27
                                   SL Seminar
                  Proposal (Cont.)
   Foreseen problems (from 3 years of experience)

    – Isolated endeavors
         No learning, poor sharing of experience (except for a few

          people)
    – Remaining doubts about the approach
    – Successful experiences not yet enough to convince people
    – Project management concepts still fuzzy for many people
    – High risk of bad project management practices (not adapted to
      people or to the size of the problems)




    28/11/97             B.Denis, M.Vanden Eynden - CERN              28
                                    SL Seminar
                           The Project
Make project-based management happen … launch a project
 Project definition:
   – Purpose
              Make project-based management happen
    – Scope
              ?
    – Objectives
              Define project procedure and practices
              Define the required organization
              Define the required support for project management

   Project organization
    – Project team members = project leaders of candidate pilot projects

    28/11/97                  B.Denis, M.Vanden Eynden - CERN       29
                                         SL Seminar
                 The Project (Cont.)
   Project plan                         Exploratory phase
                                 Identify criteria to assess efficiency
                               Define common practices and procedures
                                        Experiment approach
                                   Assess the organizational impact


     Review approach and assess efficiency

         Approval from management

                                    Implementation phase
                                       Implement support
                                    (Forum, training, guides)
                              Adapt organization (matrix approach ?)
    28/11/97               B.Denis, M.Vanden Eynden - CERN                30
                                      SL Seminar
                    By the Way ...
   “As far as possible the scientific program should be divided
    into projects with finite lifetimes and defined objectives or
    end points. These may include some long-term strategic
    projects and can be of a wide variety of sizes. There should be
    clear procedures for establishing, reviewing and terminating
    projects.”


                                               Geneva, December 1987
                                               Final Report of the CERN
                                               Review Committee
                                               (A.Abragam et al.)



    28/11/97            B.Denis, M.Vanden Eynden - CERN                   31
                                   SL Seminar
                      References
          ESA software engineering standards, ESA BSSC, ESA
           Publications Division, 1991
          Goal Directed Project Management (2nd edition),
           Andersen et al., Kogan Page, 1995
          Project Management As If People Mattered, R.J.
           Graham, 1989
          The handbook of Project-Based Management, J.R.
           Turner, McGraw Hill, 1993




28/11/97                B.Denis, M.Vanden Eynden - CERN      32
                                   SL Seminar

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:11
posted:3/1/2012
language:
pages:32