An Introduction to the Theory of Constraints breakthrough
An Introduction to the Theory of Constraints - breakthrough solution for Project Management Presenter Robert Bolton MS Project Users Group 8 th February 2000 © Avraham Y. Goldratt Institute 1998 1
Objective of this session ? Examine the effects of the problems we are facing in projects & Project Management today ? Understand the source of these problems ? Understand how TOC addresses the source of these problems ? Results to date ? Aim of TOC: Complete projects on or before the scheduled due date. © Avraham Y. Goldratt Institute 1998 2
The nature of Projects? ? Anyone ever have or heard of a project taking longer than scheduled? ? Anyone ever have or heard of a project going over budget? ? Anyone ever have or heard of a project cutting specifications or scope? © Avraham Y. Goldratt Institute 1998 3
Why is it so difficult to manage projects to deliver on time, within budget and with the full specifications/scope intact? ? Unforeseeable difficulties with vendors who supply equipment ? Longer than expected in meeting Government and/or regulatory requirements ? Unrealistic schedule ? Unreliable (but cheaper) vendors or contractors ? Unforseen emergencies ? Difficulties in matching skilled resources with project need. ? Etc What are the resulting effects ? © Avraham Y. Goldratt Institute 1998 4
What is the current experience? Standish Group Statistics (IT Projects) w 30% of projects cancelled before finished 75% of completed projects are late Average cost overruns of 189% Average time overruns of 222% w Standish Group survey results can be found at http: //www. standishgroup. com w w w © Avraham Y. Goldratt Institute 1998 5
© Avraham Y. Goldratt Institute 1998 6
Project characteristics All projects have two things in common: 1. They involve high uncertainty. 2. They involve three different and opposing commitments: Due date, budget, and content © Avraham Y. Goldratt Institute 1998 7
Could the three opposing commitments / problems be caused by the uncertainty inherent in all projects? Often projects have difficulty finishing on time Often projects have difficulty staying within budget Often scope or specifications are cut from a project Uncertainty in Projects? © Avraham Y. Goldratt Institute 1998 8
Is uncertainty really the source of all these problems? If it is, then we would never be able to find a project that had a lot of uncertainty in it that finished significantly ahead of time and that finished within budget and with specifications intact! But, there has been at least one - the U 2! It was completed in 8 months, it beat its budget and met full specs! What this means is that uncertainty is not the cause of our three problems. What cause could account for our three problems and the success of the U 2? Perhaps it has to do with the way we manage uncertainty! © Avraham Y. Goldratt Institute 1998 9
Perhaps it has to do with the way we manage uncertainty! Often projects have difficulty finishing on time Often projects have difficulty staying within budget Often scope or specifications are cut from a project THE WAY WE MANAGE UNCERTAINTY IN PROJECTS? Adding significant safety everywhere, then wasting it! © Avraham Y. Goldratt Institute 1998 10
Which time are you likely to promise? Probability of Task Duration Time 80 - 90 % 50 % A B C Your boss asks you when you can have a specific Which time are you likely to give? Probably C, project task ready. You are already busy. Your boss maybe even C+. does expect you to meet your commitments. You also In projects, with the expectation that we will give take your commitments very seriously. There may be realistic estimates of how long the task will take, some unexpected surprises (uncertainty) you will when asked, we will likely give a similar response. If have to deal with in doing the task. we have to keep our commitments and we know that With no difficulties at all, time A is a very slight our given time will be cut, we might give a time as possibility. With some surprises, time B is very likely. long as C++! If a major disaster occurs, time C is likely. © Avraham Y. Goldratt Institute 1998 11
The “student syndrome” X Murphy Task Scheduled, Available Start Date © Avraham Y. Goldratt Institute 1998 Task Scheduled Completion Date 12
The multiplying effect of multi-tasking Task A Project 1 One Week Task B Project 2 One Week Task C Project 3 One Week In order to keep each project on track, a resource does half of task A, then half of task B, then half of task C, then finishes task A, then B, then C. How long does each task take to complete? What happened to the safety time? 1/2 A 1/2 B 1/2 C How Long? © Avraham Y. Goldratt Institute 1998 13
Delays are passed on — gains are not ? Merging paths don’t allow us to benefit from tasks completed early - What’s the impact on the total project if Task A is done in only 3 days? ? What if Task C takes 8 days? ? What if Tasks A, B, and C, through some miracle, all get done in 2 days? (Will Task D be ready to start 3 days early? ) © Avraham Y. Goldratt Institute 1998 Task A 5 3 Days X Task B 5 Days Task D 10 Days Task C X 5 8 Days 14
Project Management Evaporating Cloud - Stuck between a rock and a hard place Assumption Because the best place to handle the project uncertainty is at each task Objective Requirement Prerequisite Account for uncertainty in estimating task times Add protection time to most tasks Requirement Prerequisite Meet critical timing needs (while addressing uncertainty) Don’t add protection time to most tasks Successfully manage uncertainty in projects Assumption Because there is no way to add sufficient protection that results in short enough lead times © Avraham Y. Goldratt Institute 1998 15
Let’s consider a simple project 10 10 16 20 16 16 Theoretical this project should finish within 56 days. Will it? © Avraham Y. Goldratt Institute 1998 16
How do we measure a project? A project has three elements: Completion Date - when the benefits are realised (Throughput) Scope or specifications - the definition of what is needed to be achieved so that benefits will be realised. Budget - the money invested to get the benefits Money © Avraham Y. Goldratt Institute 1998 Benefits 17
Statistics - What is the chance that the project will be completed on time? Let’s assume that each task has a probability of 80% to be finished on time. 10 10 16 20 16 16 The probability of the total project to finish on time is 26% Upper path - 0. 8*0. 8 = 51. 2% Lower path - 0. 8*0. 8 = 64% Integration: 32. 7% © Avraham Y. Goldratt Institute 1998 18
Project Management Evaporating Cloud - Adding “injections” to the reality of the conflict Objective Assumption Injection Because the best place to handle the project uncertainty is at each task The project only protects what is critical for handling uncertainty Requirement Prerequisite Account for uncertainty in estimating task times Add protection time to most tasks Requirement Prerequisite Meet critical timing needs (while addressing uncertainty) Don’t add protection time to most tasks Successfully manage uncertainty in projects © Avraham Y. Goldratt Institute 1998 Assumption Injection Because there is no way to add sufficient protection that results in short enough lead times We use aggregated buffers and shortened duration times for the task times 19
Project Planning: Direction of the Solution Have strategic protection times - Buffers 5 5 8 10 8 8 Project Buffer 28 Feeding Buffer 16 10 10 16 20 16 © Avraham Y. Goldratt Institute 1998 16 20
Change in behaviours allows aggregated buffers to be smaller than the individual safety. ie There is less Multi-tasking!!! 5 5 8 10 8 5 5 8 © Avraham Y. Goldratt Institute 1998 8 FB 8 8 10 8 Project Buffer 14 Feeding Buffer 16 Project Buffer 28 21
Does it ever happen that a resource is needed in two places at the same time? What happens if the lower path experiences uncertainty beyond its duration times and the top doesn’t? Won’t we have a need for the C Resource at the same time? A 5 B 5 C 8 E 10 D 8 C 8 Project Buffer 14 FB 8 Solution: The Critical Chain © Avraham Y. Goldratt Institute 1998 22
Identifying the Critical Chain Resource dependencies are identified once resource contention is removed. Finally, the longest path of dependent events - the Critical Chain - is identified. ( We have already removed the safety from the task times, which was half of original task estimate). A 5 D 8 © Avraham Y. Goldratt Institute 1998 B 5 C 8 E 10 23
A feasible and immune schedule Buffers - Strategic protection of the Due date A 5 B 5 FB 5 D 8 C 8 A 5 D 8 © Avraham Y. Goldratt Institute 1998 C 8 E 10 Project Buffer 17 B 5 C 8 E 10 24
Buffers are used to provide focus and early warning to protect the critical chain and due date BUFFER WATCH & PLAN ACT OK Remaining Project Buffer: 0 6 7 12 13 17 Remaining Feeding Buffer: 0 © Avraham Y. Goldratt Institute 1998 2 3 4 5 5 25
Project control - Buffer Management The mechanism for gathering data provides us a glimpse into the future so we can take action before we are in too much trouble. The organization gathers the information for the status of the buffers in the following way: Each resource that is working on the project gives a “daily” status of the time they estimate they still need to work until the task is complete. That information is used to calculate whether any buffer time would be gained or lost if these time estimates proved true. This “daily” interaction is key to reinforce new behaviors and to provide opportunities to mentor resources. © Avraham Y. Goldratt Institute 1998 26
Project Management the TOC Way - summary Project Management the TOC Way has five main components: 1. A synchronization mechanism that allows projects to be started later, but finished sooner and that clarifies resource assignment priorities. - i. e. Multi Project. 2. Planning processes that account for the needed dependencies and completion criteria. 3. Scheduling processes that concentrate safety where it will provide the most protection. 4. Changes in behavior that support a world class relay team culture. 5. Mechanisms that create “Project Control & Visibility” to assist in global decision making. Ie due dates. © Avraham Y. Goldratt Institute 1998 27
The change is both logistical and cultural! ? Building and scrutinizing individual project networks in a way that significantly reduces the opportunity for missing critical dependencies. ? Constructing project schedules in a way that recognizes both resource and path dependencies. ? Placing safety strategically to protect the project, not the individual tasks. ? Eliminating behaviors that waste safety. ? Managing resource assignments according to which project has the greatest need. (Buffer Management) ? Managing project progress according to buffer depletion - taking corrective actions when and only when required. ? Software required to support: ? Prochain Solutions - MS Project Add-on www. prochain. com ? Concerto - ERP solution Thruput. com © Avraham Y. Goldratt Institute 1998 28
Results Saturn Development Corporation Construction of new car dealerships ? ? © Avraham Y. Goldratt Institute 1998 History ? 20% overspent ? Project length 6 -9 months ? Debate over whether it was on time After implementing Critical Chain ? Within budget ? Project length 4 months ? On time per original plan/promise 29
Results Harris Semiconductor - new $250 m plant New technology product - first 8 -inch discrete power wafer fab New raw material, new automated technology New facility, doubling capacity Project scope - construction, installation, ramp-up, Focus on actual delivery of production via the Critical Chain ? ? Industry norm ? Groundbreaking to first silicon - 28 -36 months ? Time to ramp production - 18 months Harris results with Critical Chain ? Groundbreaking to first silicon - 13 months ? Time to ramp production - 21 days © Avraham Y. Goldratt Institute 1998 30
Results Israeli Aircraft Industries Wide-Body Aircraft Directorate Multi-project maintenance operation ? ? History ? Average visit per aircraft - 3 months ? Amount of work committed by customers - 2 months One year after implementing Critical Chain © Avraham Y. Goldratt Institute 1998 ? Average visit per aircraft - 2 weeks ? Amount of work committed by customers - 1 year 31
Results Balfour Beatty Civil Engineering (UK) ? 1 st Project - By pass road (A 13) 80 M GBP - 2. 5 yr ? Half way stage - well behind schedule ? Handover 2 weeks early 99% complete vs Industry norm of 80% ? ? 2 nd Project - 8 km highway (A 50)- 35 M GBP- 2. 4 yr ? Beat tender program by 9. 5 weeks ? 45 weeks earlier than contract completion date. ? Increased Project Profit Margin. © Avraham Y. Goldratt Institute 1998 32
Results - June 1999 - Elbit Systems - Israel Advanced Military Systems - 1, 900 People (50% Engineers) ? All projects management by TOC (36 projects 1 -2 years each) ? Visibility of meeting contractual milestones months ahead ? Two major platform upgrade programs met schedule within 2 weeks ? Excellent synchronisation of Program Teams based on a common language ? Recognising the fact that most of the resources in the company might be idle part of the time. © Avraham Y. Goldratt Institute 1998 33
Results - June 1999 - Seabridge - Israel Telecommunications start-up - with Multi-Service Access Platform (50 people - 35 in R&D) Purchased by Siemens & New. Bridge in November 1997 ? ? To deliver Release 2. 0 by March 1998 - Bonus tied ? Performance - 5 months late CEO reads Critical Chain - decides to implement Sept 1998 ? 1 st project on schedule; 2 nd project 2 months early; 3 rd on track to meet “Unrealistic time constraint”; ? Major review by Siemens on company turnaround © Avraham Y. Goldratt Institute 1998 34
You have been experiencing an application of Theory of Constraints TOC is the ability to construct and communicate common sense solutions Developed by Dr Eli Goldratt and The Goldratt Institute © Avraham Y. Goldratt Institute 1998 35
For further information, books & details of services to improve your projects please contact: - Robert Bolton Associate of the Goldratt Institute Probative Solution Pty Ltd Tel: 02 -9357 1455 Fax: 02 -9357 1499 Mobile: 0412 -235 616 e-mail: probativ@ozemail. com. au or Visit web site: www. goldratt. com © Avraham Y. Goldratt Institute 1998 36
- Slides: 36