Finishing Projects Fast James R. Burns Professor of Operations Management and Information Technology Texas Tech University

Next Student Chapter PMI: April 25, Rawls--NW212, 7 pm Speaker is Sandy Williams She is a former project and program manager with U.S. West She will talk about her CMMI implementation experience Her talk will be 45 min

Presentation by James R. Burns Outline--Sources Generalities Goldratt concepts Mascitelli concepts McCONNELL concepts Kerzner concepts Maturity concepts Other sources Presentation by James R. Burns

Presentation by James R. Burns Goal: Make some suggestions as to how projects can be completed fast and frugally How about you…what you suggest? Presentation by James R. Burns

Presentation by James R. Burns Most firms Recognize project management to be a core competence today Have established project management centers of excellence for training and development of project managers and project management careers Encourage their employees to propose project initiatives with simple one-page statements of work Presentation by James R. Burns

Rationale for Reducing Project Duration Time Is Money: Cost-Time Tradeoffs Reducing the time of a critical activity usually incurs additional direct costs. Cost-time solutions focus on reducing (crashing) activities on the critical path to shorten overall duration of the project. Reasons for imposed project duration dates: Time-to-market pressures Unforeseen delays Incentive contracts (bonuses for early completion) Imposed deadlines and contract commitments Overhead and public goodwill costs Pressure to move resources to other projects

Options for Accelerating Project Completion Resources Not Constrained Adding resources Outsourcing project work Scheduling overtime Establishing a core project team Do it twice—fast and then correctly Resources Constrained Fast-tracking Critical-chain concepts Reducing project scope Compromise quality (not really)

Explanation of Project Costs Project Indirect Costs Costs that cannot be associated with any particular work package or project activity. Supervision, administration, consultants, and interest Costs that vary (increase) with time. Reducing project time directly reduces indirect costs.

Explanation of Project Costs Project Direct Costs Normal costs that can be assigned directly to a specific work package or project activity. Labor, materials, equipment, and subcontractors Crashing activities increases direct costs. Presentation by James R. Burns

Practical Considerations Using the Project Cost–Duration Graph Crash Times Linearity Assumption Choice of Activities to Crash Revisited Time Reduction Decisions and Sensitivity

Things you can do when you have unlimited resources When cost is not a factor…..in the planning phase preferably Add resources Recall the n(n-1)/2 rule for comm channels Use Overtime No additional comm overhead; no additional training Subcontract work out Fast-tracking Start tasks sooner Crash Recall Brook’s Law: adding manpower to a late project only makes it later Adding resources also results in more training and communication costs and time. One advantage of overtime is that it does not entail additional training or communication. Subcontracting makes sense when there is a separate parcel of work that can be done relatively autonomously…and the subcontractor has the specialized expertise to do the work. Presentation by James R. Burns

Strategies for shortening projects when resources are limited Reduce scope Scrub requirements—requires prioritization Use critical chain concepts You cannot start doing this in the middle of a project—recall that there is significant training involved Compromise on quality—not good!! Less requirements gathering and modeling, less prototyping, less testing Presentation by James R. Burns

Notes on shortening project durations (Most of this must be done in the Planning phase) Checking for parallelism opportunities in the schedule Pull as much work off of the critical path as you can Be aware of critical chain issues Presentation by James R. Burns

More Tips on shortening project durations REUSE, REUSE, REUSE Do it right the first time Eliminate non-value-added work activities Make projects lean Avoid changes to requirements But what if the requirements are unstable?? Presentation by James R. Burns

Presentation by James R. Burns Knowledge Reuse… Requirements Reuse Based on a Classification of projects Mapped/Programmed Projects--Everything is driven by and proceeds from the requirements Project Plan Functional Specification Design Document Code Tests and Test Documentation ALL OF WHICH CAN BE REUSED Presentation by James R. Burns

The Quality View on FAST projects: Do it Right the First Time!!! The further down the lifecycle the defects are found, the more expensive and time consuming they are to fix. Presentation by James R. Burns

Presentation by James R. Burns Fast tracking It depends on how you batch the work Presentation by James R. Burns

Presentation by James R. Burns Activity A Activity B Activity C 2 4 6 8 10 12 Activity A Activity B Activity C 2 4 6 8 10 12 Presentation by James R. Burns

Activity A Activity B Activity C 2 4 6 8 10 12 Activity A Activity B Activity C 2 4 6 8 10 12

The problem of Complexity In the early days of simpler code, it used to take a day or less to fix a bug Now, with greatly increased code complexity, it takes weeks sometimes. Complexity greatly increases time and cost Presentation by James R. Burns

Avoid changes to requirements If possible, freeze requirements during the executing phase Presentation by James R. Burns

Lean Project Management Customer-perceived value should drive everything What is their value proposition?? If we were to advertise in the WSJ that we have twice as many walkthroughs as our closest competition, would that garner any additional customers for us? Remove what does not add value Presentation by James R. Burns

Time Batching--Another Time Waster Analysis paralysis Approval cycles Formal document release Regularly scheduled meetings Planning cycles Work queues Presentation by James R. Burns

More techniques for shortening projects Scrub the requirements during or prior to the planning phase Remove from the requirements those items that add little or no value Remember the Pareto principle—80% of the value comes from 20% of the functionality REMOVE SAFETY—GOLDRATT Resist multitasking and student syndrome Presentation by James R. Burns

Presentation by James R. Burns Safety Extra time placed in an estimated task time Remove safety and put it in a time buffer at the end of the project Safety, when its buried in the tasks of the project, is a bad thing because of…. Multitasking, also a bad thing Student syndrome Task dependencies Can’t be passed along or accumulated Presentation by James R. Burns

Everybody overestimates the time required to do their task According to Goldratt (This is called SAFETY, as we said) Does anybody want to talk about how much safety they put into their estimates? Is this true in software development? It is if you have an expert doing the estimating, who really knows how long it will take him Presentation by James R. Burns

What happens after that--a possible scenario The team leader adds safety time to the task to cover his responsibilities The project leader adds more safety time The project manager may add still more safety time Presentation by James R. Burns

Implication>>> Most of the time we have built into our projects is ….. Safety Safety Safety Presentation by James R. Burns

It is hard to stay focused when: There are too many project paths on-going, in parallel There are many critical or near critical paths There are many projects being managed concurrently Presentation by James R. Burns

Measurements are a major problem with projects Measurements should induce the parts to do what is good for the system as a whole Measurements should direct managers to the point that needs their attention So often it occurs that we measure the wrong thing. The wrong measure leads to the wrong behavior Tell me how you measure me and I will show you how I behave Presentation by James R. Burns

More Measurements -- EVA -5 -5 -5 +15 Presentation by James R. Burns

Projects are like chains Each task in sequence is a link in a chain Each link has two things weight, to which cost is analogous strength, to which throughput (time) is analogous Presentation by James R. Burns

Presentation by James R. Burns Cost vs Throughput Goldratt maintains that management in the cost world is a mirage efficiency becomes paramount local improvements are necessary to get global ones Goldratt suggests the managers should manage in the throughput world, a totally different paradigm must find the constraint--the weakest link concentrate on that By the way, what is the ultimate constraint??? Presentation by James R. Burns

Remember the five steps of TOC IDENTIFY the project constraint--the critical path Decide how to EXPLOIT that constraint SUBORDINATE everything to that decision ELEVATE the systems’ constraint Go back to step 1, and find another constraint Presentation by James R. Burns

Probabilistic task durations Late finishes tend to accumulate and may increase the length of the project Early finishes do not show up This explains why safety disappears Presentation by James R. Burns

Other problems with safety Is wasted by the “student syndrome” Basically, this is procrastination Is wasted by multitasking (a person who works on several tasks at the same time) With each change of task, a set up is required Is wasted by dependencies between steps These dependencies cause delays to accumulate, but advances are wasted Delays get passed on; advances don’t Presentation by James R. Burns

Problems other than safety Early start vs. late start Existing measurements are worthless because they are based on a cost world mentality, according to Goldratt Existing measurements (Earned Value Analysis) do not take into consideration the critical path We’re talking about BCWP, BCWS, ACWP, CV, SV, CPI, SPI, BAC, EAC, etc. Presentation by James R. Burns

Early Start vs. Late Start B 5 A 5 E 10 D 10 C 10 Presentation by James R. Burns

Presentation by James R. Burns Solutions Take the safety out of the individual tasks and put it at the end of the critical path in the time buffer, called a project buffer This means making the tasks roughly 50% as long as they would otherwise be. Presentation by James R. Burns

Presentation by James R. Burns More solutions At the point where each feeding path intersects with the critical path, place another time buffer, called a feeding buffer. The feeding buffer protects the critical path from delays occurring in the corresponding non-critical paths. When resources are needed on the critical path, these resources are advised ahead of time exactly when they must make themselves available. When that time comes, they must drop everything else and do the required critical tasks. Presentation by James R. Burns

Measurement solutions Measure progress only on the critical path; what percent of the critical path we have already completed. This is all we care about!! Have a project or team leaders measure progress on a non critical paths in terms of unused feeding buffer days Presentation by James R. Burns

Shrinking the task time: Effects There is less procrastination There is much more focus There is less multitasking Presentation by James R. Burns

Presentation by James R. Burns More Suggestions Put your “BEST” people on the critical path Watch out for critical chains Presentation by James R. Burns

Presentation by James R. Burns What are the ramifications of a delayed software product, intended for commercial sale? Less market share Less profit; maybe no profit Lower analyst profit expectations Declining share price Out of business? How many firms has Microsoft driven out of business? Ask Philippe Khan (founder of Borland) what the implications of getting a product late to the marketplace are Presentation by James R. Burns

What about Procurement Most firms enter into LOSE/LOSE Strategies A fixed-price lowest bidder contract is LOSE/LOSE Strategy This forces Contractors to under bid their costs, hoping to make it back on the changes to the requirements that the customer will have to pay for Instead, Contractors should be induced to deliver product on time, with as much functionality as possible How would you do this? Presentation by James R. Burns