CfAO Fall Retreat Career Development Sessions. Project Management. Dave Palmer Mike Fitzgerald Lisa Hunter. Lake Arrowhead, November 6 th - 9 th, 2008

Transcription

1 CfAO Fall Retreat Career Development Sessions Project Management Dave Palmer Mike Fitzgerald Lisa Hunter Lake Arrowhead, November 6 th - 9 th, 2008 LLNL-PRES

2 Session Outline Introduction Project Organization Presentation / Activity Project Requirements Presentation / Activity Project Planning Presentation / Activity Project Tracking Presentation Risk / Contingency Management Presentation / Activity Series of Unfortunate Events Activity Follow-up on Unfortunate Events Discussion Management & Proposal Writing Presentation Geographically Diverse Teams Presentation Management Styles Presentation / Discussion Session Feedback 2

3 Introduction, 1 Project Management is the planning, organizing, and control of a collection of interrelated tasks to create a unique product, service, or result. A project is a finite endeavor having specific start and completion dates undertaken to create a quantifiable deliverable. Projects undergo progressive elaboration by developing in steps and predictable increments that are tied to benchmarks, milestones, and completion dates. ID Task Name Qtr 4 Qtr 1 Qtr 2 Qtr 3 Qtr 4 Qtr 1 Qtr 2 Qtr 3 Qtr 4 Qtr 1 Qtr 2 5/30 1 CDR 11/18 2 Start AO SATP 1/27 3 AO SAT Rev iew (will hav e to rev iew components, not 4 OMSS arrives at UCSC 2/1 5 6 procure parts construct testbed testing 18 UCSC 19 MEMS - Engineering 20 MEMS - Science 21 LLNL 22 AO algorithms with generic computer 23 WFS camera 24 WFS 25 AO Computer (AOC) 26 T/T / woofer 27 MEMS 28 HIA 29 T/T / woofer integrate into OMSS and test electrical (LLNL) software (LLNL) 3 (text from Wikipedia, Project management, mod.)

4 Introduction, 2 Project managers must manage: People (and other resources) Time (schedule) Money (budget) Project scope 4 (text from About.com, Project Management 101, F. John Reh, mod. figure from

5 Project Organization Disorganized Project What s wrong with this picture? 5

6 Project Organization Small Project A saner and more workable organization: Sponsor Principal Investigator Resource Analyst Project Manager Engineer 1 Scientist 1 Engineer 2 Technician 1 6

7 7 Project Organization Medium Project

8 Project Requirements When people talk about requirements, they re usually talking about technical requirements, such as: how big the widget that you re designing and building can be and what it can weigh how much power the widget can draw what environmental conditions it has to survive in what its inputs and outputs will be and what the quality of these need to be 8 But, there are also management requirements: how long you can take to design and build your widget how much money you can spend designing and building your widget whether you re expected to maintain a certain level of contingency what your reporting requirements are what deliverables there are besides the widget, itself (e.g., documentation)

9 Project Planning Schedule, 1 9 The project schedule is, of course, the thing that says what tasks will happen when. To avoid trying to squeeze the project into the available time and money, do a bottom-up schedule early-on without worrying about how much time or money you actually expect to have. To do this: work with people representative of those who will do the work to break the project into tasks and sub-tasks (doing a Work Breakdown Structure (WBS) can help here) have your representative people estimate the amount of Work (person-hours) required to accomplish the subtasks this is normally done using history (previous experience with similar tasks) or, for commonly done tasks, standards given your available resources (people), determine how much calendar time the subtasks will take remember that n people usually can t get a task done in 1/n time figure out which tasks depend on which other tasks; in other words, which tasks need to be done in series use a scheduling program (like Microsoft Project) or paper (for a small project) to form this all into a overall schedule set project milestones (like reviews, etc.)

10 Project Planning Schedule, 2 Projects are very commonly divided into phases: Conceptual Design Preliminary Design (~20%) Critical (or Detailed) Design (~20%) Build (~40%) Test (~20%) Commissioning ID WBS Ta Dur 1 1 Deve 40wks Prelimi 8wks Critic 8w Build 16wk Test 8wks Phases are commonly concluded with reviews: Qtr4 Qtr1 Qtr2 Qtr3 Qtr4 40wk Conceptual Design Conceptual Design Review (CoDR) Preliminary Design Preliminary Design Review (PDR) Critical Design Critical Design Review (CDR) Build Pre-test Review Test Pre-ship Review 8wks 8wks 16wks 8wks 10

11 Project Planning WBS A Work Breakdown Structure (WBS) can be very helpful in identifying levels of deliverables and, then, the tasks associated with those deliverables. A few notes: the WBS should be complete, including all work required for the project (so, any work not in the WBS is out-of-scope) WBS units are called Work Packages Work Packages: are the lowest level WBS elements have an accompanying narrative have three measurable components: scope of work, total cost, and timeframe 11 (figure from PMI's Practice Standard for Work Breakdown Structures (2nd Edition))

12 Project Planning Gantt Charts Gantt charts simply illustrate project schedules. Notice: subtask starts and finishes and rolled-up tasks milestone dependencies work complete 12 ID WBS T AOS Build/ receiv WFS WFSca WFS(f twete AO C T/ t twete twete WFSc WFS T/T/ s co eval refine r e GPIsp algori opto-m A Dev I&Tan 3rdQua 4thQuarter 1stQua 2ndQu 3rdQ 4thQ May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov 8/1 10/1 12/1 2/27 3/2 3/4 3/4 4/27 5/25 7/8 7/17 7/2 8/4 8/4 8/5 8/14 9/16 10/1 1/6 1/1 1/6 1/1 1/1 1/1

13 Project Planning Resources Determine resources needed to perform the project resources are the people and facilities needed to perform the project: People scientists, engineers, technicians, managers, financial analysts, IT support, purchasing agents, administrators, Office space, computers, printers, copiers Lab space, equipment, and supplies Misc. consumables (e.g., office supplies), teleconferencing, 13

14 Project Planning Budget A project budget is composed of: labor costs the sum of the cost, per hour, for each individual that is working on the project times the number of hours that he or she is scheduled on the project facilities costs (these are often covered by your institution and/or included in labor costs) costs for goods and services to build whatever you re building or do whatever you re doing these should be as accurate as possible: list prices for Commercial Off-The-Shelf (COTS) products estimates (good), budgetary quotes (better), or hard quotes (best) for semi-custom or custom goods and services travel costs misc. costs (e.g., IT support, office supplies, etc.) 14

15 Project Planning Microsoft Project Microsoft Project, or a similar program, is a relatively convenient (but, nowhere near painless) way to do a lot of what we just talked about. I ll show you an example 15

16 Project Planning Activity Now that I ve shown you Microsoft Project, we re not going to use it. Instead, we re going to use good old pens and paper to: identify the tasks and subtasks required to execute your project, and who will be performing the tasks (do a quick WBS, if you want to) schedule the tasks identify any other resources required to execute your project (assume your department will cover office and lab space) prepare a budget The deliverables for this activity are: a Gantt chart a budget Reminders: on Gantt charts, triangles indicate task starts and finishes, diamonds indicate milestones remember to leave some schedule and budget contingency 16

17 Project Tracking Reviews In a perfect world, the following would be demonstrated / completed for each review: Conceptual Design Review (CoDR) the project is feasible no laws of physics will be broken; no aspect of the project will cost too much or take too long Preliminary Design Review (PDR) all requirements have been fully defined and flowed-down; all interfaces have been fully defined; all models / simulations have been performed to demonstrate performance; virtual and/or physical prototypes have been constructed to assure proper fits Critical Design Review (CDR) all detailed design work has been completed and documented (fab drawings, schematics, software detailed designs, etc.); the project is ready to cut metal ; there are final draft for all test plans Pre-test Review all Build phase tasks have been completed and the Acceptance Test Plan (ATP) is sufficient to test all requirements 17 Pre-ship Review the project has been shown to fully meet its requirements by successfully completing its ATP

18 Project Tracking EVA, 1 Earned Value Analysis (EVA; a.k.a. Earned Value Management (EVM)), is an effective way to track past performance and project future performance. time is money EVA works by converting time into money to give the following: Budgeted Cost of Work Schedule (BCWS) this is the cost of the work that should have been performed by today s date, as per the project plan, for a given task (equals Baseline Cost times fraction of time that has elapsed) Budgeted Cost of Work Performed (BCWP) this is the cost of work that has been performed by today s date, as per the project plan, for the given task (equals Baseline Cost times % Complete / 100) 18 Actual Cost of Work Performed (ACWP) this it the amount of money that has actually been spent by today s date, for the given task (equals Actual Cost)

19 Project Tracking EVA, 2 Tracking inputs: % Complete this is an estimate of the percentage of a given task that has been completed once the plan has been established, this is the most subjective input (beware of garbage in, garbage out ) Actual Cost this is the amount of money that has actually been spent on a task normally, this would come from a financial report Project Trends Derived metrics (for tracking / projecting): Cost variance = BCWP - ACWP Schedule variance = BCWP - BCWS = 0 means on schedule / budget < 0 means behind schedule / budget > 0 means ahead of schedule / budget Dec Jan Feb Mar Apr May SPI CPI CSI Cost Performance Index (CPI) = BCWP / ACWP Schedule Performance Index (SPI) = BCWP / BCWS = 1 means on schedule / budget < 1 means behind schedule / budget > 1 means ahead of schedule / budget 19 Cost Schedule Index (CSI) = CPI * SPI further from 1.0 means less likely for project to recover

20 Project Tracking Periodic Reports Many sponsors require periodic reports, usually either monthly or quarterly. Reports generally contain: Status summary Technical status narrative status description by discipline (mechanical, electrical, software, etc.) and/or by subsystem technical risk status Management status narrative status description schedule status (what s ahead and what s behind) budget report (how much money has been spent on what (e.g., labor, procurements, travel, misc) EVA management risk status Conclusion 20 It s also quite common to have reports supplemented with periodic reviews, such as quarterly or annually (for big projects).

21 Risk / Contingency Management, 1 Risk management is one of the most challenging aspects of project management, because it means trying to predict the unpredictable. Potential risks should be identified and characterized early, the first day a project is considered is not too soon! 21 Although there are different ways to characterize risks, a common way is to: estimate the budget and schedule cost for each risk (for example, if a particular risk occurs, it might cost the project $10k and 1 week that aren t currently in the project plan) assign each risk a Probability of occurrence grade (1 -> 4, for example) and a Consequence of occurrence grade (also 1 -> 4) combine these to produce a weighting = Probability * Consequence / 16 apply each risk s weight to its estimated costs to give weighted budget and schedule costs sum the weighted budget costs and sum the weighted schedule costs to give the project s estimated risk exposure perform these steps periodically to keep the project s estimated risk exposure updated

22 Risk / Contingency Management, 2 Other aspects of risk characterization: Are there mitigations for risks? What are they? Can you over-design something, if necessary, to compensate for fuzzy requirements? Can you mitigate schedule risk by doing some work ahead of time, even though you might not end up needing it? Can you set aside some money, in case something costs more than expected? 22 Decision points should be associated with risks On what basis will you be able to either retire a risk or deal with it and when do you expect to be able to decide that? Are there intermediate dates when you will have more information that will let you downgrade a risk or formulate new mitigation strategies? For complicated risks, a risk tree can help map out decision points and strategies. The dates associated with decision points can be added to your project schedule as milestones, or maintained separately.

23 Risk / Contingency Management, 3 A very important risk mitigation for a project is contingency. There are 3 general kinds of contingency: budget contingency money set aside, with no specific purpose to begin with (30% budget contingency is not an unreasonable number for a challenging development project) schedule contingency extra time at the end of a project, or at the end of tasks or phases, to accommodate work taking longer than expected scope contingency decreasing requirements to compensate for: unexpected technical challenges, being over budget, or being behind schedule A rigorous contingency analysis involves applying confidence levels to schedule and budget items and then rolling up the results. 23

24 Risk / Contingency Activity Now we re going to identify risks and assign contingencies for your projects: identify at least 5 risks for your project estimate the budget and schedule cost for each risk assign each risk a Probability of occurrence grade (1 -> 4, zero means no risk) and a Consequence of occurrence grade (also 1 -> 4) produce a weighting for each risk = Probability * Consequence / 16 apply each risk s weight to its estimated costs to give weighted budget and schedule costs for this activity, these are your allocated contingencies sum your allocated budget and schedule contingencies and subtract these from the contingencies that you set aside when we did the first project planning activity this gives your unallocated contingency (which may be negative; don t worry about that, for now) The deliverables for this activity are: 5 or more graded risks for your project allocated budget and schedule contingencies for each risk your unallocated budget and schedule contingency amounts 24 After we re done with this, we re done planning

25 Series of Unfortunate Events Now, we re going to run through the project that you ve been planning in real-time. Rules of the game: several months will be scaled down to about 30 minutes unless otherwise indicated, time and money will be spent as per your plan if you have anticipated an unfortunate event and have contingency for it, use the contingency; otherwise, improvising is allowed (but be fair) if you run out of time or money, consult with your sponsor (me) 25 the goal is to see how close to on-schedule and on-budget you end up; but, don t worry about being super accurate a more important goal is to think about how you would handle real world events (and be ready for them next time!)

26 Follow-up on Unfortunate Events How many groups came out pretty much on-budget and onschedule? Any under budget and ahead of schedule? How did different groups handle the unfortunate events? Would you plan your project differently now? If so, how? 26

27 27 Management & Proposal Writing Most proposal writers focus on the technical aspects of a project. But, the project management aspects are just as important. In fact, in this day of greater competition and tighter budgets, a technically perfect proposal is still likely to lose if the team doesn t demonstrate management expertise and a solid management plan! One other point balancing realistic estimates with winning the project is difficult: if you re overly optimistic, you end up behind schedule and over budget and, possibly, with a canceled project if you re overly pessimistic, you might lose a project that you might have won, otherwise

28 Geographically Diverse Teams Geographically diverse teams are becoming more common. This is a little different than a prime with several subcontractors, in that the teams are much more interactive with each other. 28 A few tips: know and respect the culture of your teammates try to keep the institutions on the team collaborative rather than competitive make sure you have the necessary tools for frequent interaction: teleconferencing capability videoconferencing capability a central repository for documents (like DocuShare or KnowledgeTree lots of travel money in your budget use the tools (a lot)

29 Management Styles Dictatorial manager takes no input from individuals what he or she says goes with no ifs, ands, or buts. Authoritarian manager asks for and accepts suggestions from individuals; but, may or may not use those suggests to make his or her decisions that are then, generally, non-negotiable. Participative manager actively encourages inputs from individuals and involves them in the decision-making process; but, he or she still has the last word, if necessary. Laissez-Faire manager lets the group and/or individuals make their own decisions; will step in if a decision cannot be made, there are conflicting decisions, or other problems arise. Which styles would work best for which management situations? 29 Where do you feel that you fit in this spectrum?

30 Session Feedback We encourage both written and verbal feedback: Please take a few minutes to fill out the evaluation form that will be handed out and, then Stick around for verbal feedback Thanks! 30