CaltransProjectManagementImprovementProcess

1、Caltrans Project Management Improvement ProcessRevised July 11, 2003Page #Caltrans Project Management Improvement ProcessRevised April 22, 20061. RESEARCH AND RECOMMENDATIONS 11.1 Initiate - Project Charter 11.2 Plan 21.3 Research 31.4 Identification of Improvement Needs 41.5 Recommendations 51.6 Co

2、ntrol 81.7 Close 82. DEVELOPMENT AND IMPLEMENTATION 82.1 Plan -See 1.5.282.2 Manual, Handbook or Guide92.3 Minor Tools 102.4 Training113. EVALUATION: Does the product meet the criteria for success listed in the Charter?11APPENDICES 12Appendix A: The Caltrans Project Management Improvement Process 12

3、Appendix B: Standard outline for a Project Management Improvement Charter 13Appendix C: The Caltrans Project Management Improvement Workplan14Appendix D: SPMIT Review Process17Appendix E: Standard Amendment Form18Appendix F: Sample Process Description19Appendix G: Process Description Form 20Appendix

4、 H: Project Management Roles 23This document describes the Caltrans Project Management Improvement Process. Readers are advised to begin by referring to the process flow diagram in Appendix A. Many readers will be able to begin the process using only Appendix A. You are advised, however, to refer to

5、 this document before starting each of the activities shown in Appendix A. The text of this document provides additional supporting information that may not be immediately obvious from Appendix A.1. RESEARCH AND RECOMMENDATIONS1.1 Initiate - Project CharterSection 5.1.3.1 ofa Guide to the Project Ma

6、nagement Body of Knowledge(the P“MBOK?Guide” ), which has been adopted as the Caltrans standard, lishtes etssential requirements for a Charter. It says, The project charter should be issued by a manager external to the project and at a level appropriate to the needs of the project. It provides the p

7、roject manager with the authority to apply organizational resources to project activities. It continues, It should include, either directly or by reference to other documents:? The business need that the project was undertaken to address.? The product description.The main goal of a charter is to ens

8、ure that the project manager understands what the sponsor wants. Once the project manager is confident that he or she understands the sponsors written expectations, the team should proceed with project planning and execution.Every Project Management Improvement charter shall follow the outline shown

9、 in Appendix B.The Project Manager, appointed by an Executive Review Committee (ERC), prepares the charter.The Project Manager leads a project team, which consists of:? The Project Manager? A Division of Project Management representative, who provides staff support to the team.? Subject Matter Exper

10、ts from Districts, Regions and Headquarters.The ERC and the Chief, Division of Project Management approve the Charte. rThe Division Chief s signature marks the completion of Milestone A“ Division Chief approves ProjectCharter ”.1.2 PlanFigure 3.5 of the PMBOK? Guide outlines the generally accepted p

11、rocess for developing a project plan. This process begins with the development of a project charter (section 5.1 of the PMBOK? Guide ), proceeds through a more detailed scope statement (section 5.2) to the development of a project-specific Work Breakdown Structure (WBS).The PMBOK? Guide describes th

12、e development of a projects-pecific WBS as “ Scope Definition ” (Section).5T.3he WBS has two goals:1. To ensure that the project includes all the work needed2. To ensure that the project includes no unnecessary work.Appendix C provides some possible elements for a process improvement WBS. A given pr

13、oject will probably not need all the elements. The project team should ask of each element“ is thisproduct necessary for the success of the project.” No unnecessary products should be includeThe following products are almost always necessary:? A Project Charter (1.1). Without a written charter, the

14、project manager cannot be sure that he or she understands the sponsors expectations.? A Plan for each phase (1.2 and 2.1). Team members must know what they are expected to produce and when it is due.? A training plan (1.5.2.1 and 2.4). Each project will affect hundreds, perhaps thousands, of Caltran

15、s employees throughout the State. The project must include a plan for ensuring that they understand the products of the project and that they can use those products effectively.? Allowance for project change (1.6). Projects seldom, if ever, develop exactly as planned. The project team must be ready

16、to change their plans to accommodate changes while still delivering the required product on time and within the constraints set in the charter.Caltra ns Project Man ageme nt Improveme nt ProcessRevised July 11,2003Page 91.2.1 Consultant ProcurementIf a consultant is needed, add the following WBS ele

17、ments to the project plan:121.1 Con tract ing pla n1.2.1.2 Request In formati on from Con sulta nts (“ Request for Offers, ”“ Request for Prop*“ Request for Qualifications,” etc.)1.2.1.3 Con sulta nt Selectio n and Con tract Award or Task Order Approval1.2.1.4 Con tract admi nistrati on and closureC

18、on tracts are gover ned by State law, regulatio ns and the policies of the Departme nt of Gen eral Services and the Divisio n of Procureme nt and Con tracts. There are differe nt procedures for each type of con tract. They all follow the above basic four-step processes.1.3 ResearchThere are gen eral

19、ly two aspects to the research for a process improveme nt project:? Exter nal research, to determ ine how comparable orga ni zati ons perform the process and what curre nt bus in ess philosophy and tech no logy could affect the process.? Internal research, to determ ine how this orga ni zati on perf

20、orms the process.1.3.1 Documentation of the State of the Art131.1 Docume nt SearchExter nal research may in clude readi ng books, jo urn als and other docume nts and research on the In ternet.1.3.1.2 Con sultati on with ExpertsExter nal research may in clude hiri ng con sulta nts who have specialist

21、 expertise in the curre nt philosophy and tech no logy that could affect the process.1.3.1.3 Atte nd ClassesExter nal research may in clude participatio n in courses that teach the curre nt philosophy and tech no logy that could affect the process.1.3.1.4 BenchmarkingExter nal research may in clude

22、con tact ing comparable orga ni zati ons to determ ine how they perform the process.1.3.1.5 Product and Software ReviewExternal research may in clude the testi ng of tech no logical products that are used by others in the process.1.3.2 Documentation of Current State in Caltrans132.1 In terviewsInter

23、nal research will gen erally in clude in terviews of Caltra ns subject matter experts to discover how Caltra ns curre ntly performs the process.1.3.2.2 SurveysInternal research may in clude surveys of Caltra ns employees. To en sure that they will produce statistically valid results, surveys should

24、be desig ned or reviewed by people with training in market research. Surveys shall not be admi nistered without prior approval from the Caltra ns Labor Relati ons office. They should be tested on a small sample before being used on the target audie nee.1.3.2.3 StatisticsInternal research may include

25、 the gathering of current Caltrans statistics. This may include acco unting data, reports from other Caltra ns databases, etc.1.3.2.4 Curre nt framework and flow chartsCaltra ns has aCaltra ns Project Man ageme nt Framework This con sists of a detailed diagrammatic breakdow n of Caltra ns project an

26、d program man ageme nt processes and a flowchart show ing the lin kages betwee n those processes. Most project man ageme nt process improveme nt projects will in clude a review of the releva nt sect ions of theFramework.1.4 Identification of Improvement Needs1.4.1 Determ ine desired state1.4.1.1 Vis

27、io n stateme ntOnce the research is complete (1.3), the team should agree on a desired state for Caltrans. This may begin from a comparison of the“ State of the Art ” (1.3.1) against the current Caltransprocess (1.3.2). This is one of the most critical and creative parts of the project. It requires

28、that team members give up any pre-c on ceived no ti ons and env isi on what Caltra ns could become.They should beware of slavishly following fashion the“ State of the Art ” might not be the“ Desired State. ”The “ desired state ” should be encapsulated insiobrisifetement. This should be as short as p

29、ossible. The Ion ger the stateme nt, the more likely that decisio n-makers will be distracted by subord in ate clauses and the less likely that there will be agreeme nt.1.4.1.2 Spon sor con curs with visi onHaving determined their“ desired state, ” the Project Manager must obtain approval from theER

30、C, the Divisio n Chief and other affected decisi on makers. During this process, the“ desiredstate ” will probably change. The project manager and team must work quicniy flexibly to address the concerns of the reviewers. These concerns can ofte n be addressed by minor cha nges in wording. The projec

31、t manager should take care to explain that the goal at this stage is only to agree on a visi on. Impleme ntati on comes later.1.4.2 Gap an alysisGap an alysis is a comparis on betwee n the approved visi on and the curre nt state. The team should list what aspects of the current process do not match

32、the vision.1.4.3 Most likely solutio nsThis is another creative part of the project. The team needs to agree on the solutions that aremost likely to succeed in address ing the deficie ncies. A possible process for doing this is:? List the deficie ncies (from 1.4.2)? Through a voti ng system, ra nk t

33、he deficie ncies (e.g., each team member has five votes that they must assig n to at least three differe nt deficie ncies. Team members should not assig n all their votes to only one deficiency. If the rest of the team does not think that deficiency is most sig ni fica nt, they could be“ losysif on

34、eSiromber focuses on only onedeficiency, they might skew the results and produce a“ highest rank ” that nobody elssupports.)? Brai nstorm possible soluti ons for the highest-ra nked deficie ncy. Gen erally, if one solves the largest problem, some lesser problems will be solved along the way.? Discus

35、s pros and cons of each soluti on. Con sider which soluti ons will best address the deficie ncy withi n the resources that are available or could reas on ably be made available.? Use the voti ng system to rank the soluti ons.? Repeat for the sec on d-ra nked deficie ncy and other deficie ncies that

36、ran ked n ear the top of the list.1.4.4 Proposed framework and flow chartsThe team should con sider how their most likely soluti ons would cha nge thFramework described in 1.3.2.4. They should prepare a revisio n of their porti on of thFramework and submit an ame ndme nt to thbramework.1.4.5 PilotTh

37、e team should test the adequacy of the proposed process developed in 1.4.4. This is the DoCheck part of the sta ndard quality improveme nt cycle (Pla n-Do-Check-Act)1. The test may bean actual use of the process on a State Highway project or a role-play of the actual use. After the pilot, the team s

38、hould revise the process developed in 1.4.4, if this is n eeded.1.5 Recomme ndatio ns1.5.1 Draft Process and, possibly, Draft DirectiveThe team shall develop a recomme nded process. It may recomme nd a Project Man ageme nt Directive, provided that it does not duplicate any part of an existing direct

39、ive.Walter A. Shewhart first discussed the concept of PDCA in his 1939 book, Statistical Method from the Viewpoint of Quality Con trol , although it was Shewarts prot g eDem in gMho promoted the now widely recog ni zed four step process for con tin ual improveme nt. Demi ng refers to it as the PDSA

40、Cycle (Pla n Do Study Act) or the Shewhart Cycle. Others call it the Demi ng Wheel.1.5.1.1 ProcessesThese are n arrative descripti ons of the framework that was developed in 1.4.4. Processes follow a SIPOC model, diagrammed below； The n arrative should describe what the in puts are, who supplies the

41、m, what process is performed, who performs it, whe n they perform it, what the output is and who the customer is of the output.jpplierI nput*rocess-O utput*The ERC approves the process. It should also be submitted to the Statewide ProjectManagement Improvement Team for review (see Appendix D). The D

42、ivision Chief s approvalmarks the completi on of Milestore B “ Divisio n Chief approves Proposed Process”.For each task on the proposed framework provided in 1.4.4, the team should produce:1. Task Title2. Task Description3. In puts to the task, together with where the in put comes from. If it comes

43、from ano ther project man ageme nt task, check that the in put is listed as an output of that task.4. Outputs from the task, together with where the output will be used. If it is used by another project man ageme nt task, check that the output is listed as an in put to that process.5. Description of

44、 the Procedures and Techniques used to perform the task. If outputs are produced only under certain conditions, the conditions must be described in precise detail. There must be no ambiguity about these con diti ons.6. Roles and responsibilities7. Tools to perform the task, such as:a) Templatesb) Ch

45、ecklistsc) Samples of completed templatesAppe ndix F shows a sample of these products and Appe ndix G is a template for completi ng the product.1.5.1.2 Draft Project Management DirectiveDirectives establish statewide minimum sta ndards for project man ageme nt. Before propos ing a directive, team me

46、mbers must review all existi ng Project Man ageme nt Directives to en sure that their proposed minimum standard is not already contained in an existing directive. The Project Man ager must en sure that each new directive does not duplicate any part of an exist ing directive. If there is a com mon ar

47、ea betwee n two directives, the later directive should refer to the earlier directive rather tha n repeat ing what has already bee n said.2 SIPOC is a widely- acclaimed method of process analysis. It is part of the -Sigma” adopted ByxMotorola Corportation in the 1980s, and it is based on the princip

48、les taught by Walter A. Shewhart and W. Edwards Deming.1.5.2 Implementation Plan1.521 Skill Developme nt Pla nEach process improveme nt requires a behavior cha nge by some Caltra ns employees. The project must therefore have a pla n to en sure that the employees un dersta nd the required cha nge and

49、 have the skills to perform successfully.The project team should con sult with the Office of Capital Project Skill Developme nt. If trai ning is needed, the Office of Capital Project Skill Development will develop that training, which may con sist or formal training or on-the-job aids.The project te

50、am rema ins resp on sible for en suri ng that the behavioral cha nge occurs throughout the State.1.5.2.2 Timeli neThe team should update their schedule to show the activities necessary to develop and implement the recomme nded soluti ons.1.5.2.3 ResourcesThe team should recomme nd who should perform

51、 each of the activities in the impleme ntati on pla n and estimate the effort invo Ived.1.5.2.4 Marketi ng Pla nTeam members should prepare a pla n to address the vital factors for tran sformatio n. Every team member must be fully en gaged. The vital factors ar$:1. Establish a sense of urge ncAt lea

52、st 75 perce nt of man agers must be genuin ely convin ced that the cha nge is n eeded.2. Build a powerful coaliti on At least 50 man agers must be committed to the cha nge.3. Define the visio n -See 1.4.1.1. The visio n stateme nt must get a react ion that in dicates both un dersta nding and in tere

53、st.4. Communi cate the visio n Use every available com muni cati on method.5. Remove obstacles - The best cha nge efforts will ofte n fail if those who are asked to cha nge cant or wont act. En sure that people have the means and the training to take acti on, and the permission to act without fear o

54、f retribution or punishment for doing so. Remember, you may think you have empowered people to act, but it is their perception of this empowerment that will, i n the end, determ ine if and how they do act.Kotter, John P., Leading Change, Harvard Business School Press, Boston, MA, 1996.6. Celebrate s

55、hort-term wins - Large change efforts take a long time. It is easy for people to lose their en thusiasm over an exte nded period. Celebrati on of short term wins helps keep people motivated and your cha nge efforts on track.7. Do not declare victory too soon - True orga ni zati onal cha nge takes Io

56、n ger tha n you thi nk. Itis tempti ng to assume, after a few mon ths, that cha nge has occurred. But like a rubber band that has been stretched only once, people will attempt to return to the old way of doing things.8. In stituti on alize the cha nge - Un til the cha nge becomes a way of life, effo

57、rts to con solidate the cha nge must con ti nue.1.6 Co ntrolProjects seldom, if ever, develop exactly as planned. The Project Manager must change the plans to accommodate changes while still delivering the required product on time and within the constraints set in the charter. It is important to cha

58、nge the plan before continuing. If one does not update the pla n, the project soon deviates so far from the pla n that the pla n becomes useless. Team members no Ion ger know what they are expected to produce and whe n it is due.1.7 CloseThe materials developed in Phase 1 should be archived and the team should con duct a“learned ” session. Archived material and lessons learned must be available for future teams. If n eeded, the Project Man ager should subm