A brief history of Scheduling

Hi Samer,

Sorry, I can’t, but if you are going to ask me about another "spider" - Spider Solitaire, then I would say that I’m an expert.

Regards

Dear Anoon,

No Solitaire :) but I will update you in good time.

With best regards,

Samer

http://www.pmforum.org/library/papers/2006/A_Brief_History_of_Scheduling.pdf

- "Conclusions
The evolution of scheduling has been a fascinating journey:
• Kelley and Walker set out to solve the time-cost conundrum and invented CPM. For most organizations the resolution of time-cost issues is still in the ‘too hard’ basket (although
SPIDER offers an interesting solution)!"

There is no mention as to the fact that few software, if any, other than Spider can provide you with true Float after resource constraining. This has been known for decades at the college level thesis but not at a practical software application until now.

Maybe the article was written before this functionality was available in Spider.

Best regards,
Rafael

Thank you, Samer,
the paper is interesting and I found it on another site http://www.mosaicprojects.com.au/PDF_Papers/P042_History%20of%20Scheduing.pdf.
It is interesting that Russian joint venture of ICL (ICL-KPO) is one of our Russian distributors.

Rafael,
Spider Project calculates Resource Critical Path and true resource constrained floats since 1993 when it appeared on the market. In 1996 I had made presentations at IPMA Congress in Paris and PMI Conference in Boston speaking about Resource Critical Path and paying special attention to the wrong total floats calculated by known software packages when resources are constrained. Since then nothing had changed except the invention of Critical Chain (other name of Resource Critical Path).

There is a lot to write about resource constrained scheduling but I expect that Patrick decided to make very brief guide.
I believe that an author of this paper is PP member.

Best Regards,
Vladimir

Dear Rafael,

Are you sure that Primavera is not calculating a resource critical path? Currently each activity has "Steps" into it. And each step can have so many weights associated with its completion and that surely affects the float on each activity.

With kind regards,

Samer

Samer,

I believe in P6 you can schedle an activity without the use of steps, this is a functionality with other purposes, but I am not a P6 user.

If you have P6 you can try with the following monstruous 2 activities job.

Schedule a 2 activities network, not tied one to each other by precedence logic but sharing a limited resource.

For example:

Activity A – 5 days
Activity B - 5 days
Both activities claiming for Resource A of which only one is availble.

Most software under resource leveling will schedule A and B at different time, the first one depending on your priorities selection. This is ok up to this point, but the problem is most will show float of 5 days on the first activity while a delay in this activity will delay the completion of your schedule.

Imagine that you as a Project Manager look at the available float as per your CPM schedule and start paying attention to those shown as critical and even start moving resources from those showing some float as to keep under control those showing no float. Because most software under resource constrained schedule will show the wrong float for resource critical activities you most probably will end up moving resources precisely from the wrong activities. You either will discover too late it was wrong or start doubting about the CPM float display. Well better start doubting about the CPM float display and either buy a glass ball or software capable of showing resource critical float.

You got to know what is driving your schedule in order to take control.

Best regards,
Rafael

The following illustrates how wrong the computations for float on a simple 2 activities job can be.



Best regards,
Rafael

Dear Rafael,

I will delay answering your question until I read the manuals. You can assign a Calender for each resource. There are so many options for each activity, and unless you understand the basic programming theory, I do not think I can answer the question correctly.

With kind regards,

Samer

Sammer,

As to keep it simple limit the schedule to 2 activities, a single resource and a single calendar, it still will be wrong. The wrong output from SureTrak was as per 2 activities, a single resource and a single calendar.

Another way to find out the computations are wrong is if some activities are missing is by filtering for critical activities and if you see a discontinuous path most probably the cheap software left some out because is calculating wrong float for resource critical tasks.

Meanwhile to make it easier I will look for a single activity sample job where the single activity is not critical, of course without setting a job finish constraint or any other dirty trick to bypass computations.

Best regards,
Rafael

Hi Rafael

In Powerproject if you have set up a resource leveling operation - which is very rarely in my case - on each reschedule a note comes up which tells you which of the levelled tasks have not been moved and have therefore affected the critical path.

Not as good as Spider I agree.

Best regards

Mike Testro

PS I’m gonna kill that pesky Gene Kelly impersonator.

Mike,

You can have a thousand activities delayed by resource constraining and none critical. That resource leveling delayed an activity does not means the activity became critical.

But just imagine my two activities schedule and adding a third activity not tied to any other, not requiring any resource and with 20 days duration. After resource leveling one of the remaining two activities will be delayed but none of these two will be critical, of course most software will still show wrong float for these activities.

Best regards,
Rafael

Vezi mai multe video din Muzica
PS. Mike if you have not seen the movie Green Zone you will find another impersonator of a PP member.

Dear Rafael,

Can you please make it a 5 activity sample with start and finish milestone, and confirm that you are scheduling.

I am finding it very hard to accept that CPM is changing from one software to another. The basic programming must be the same. In my opinion, problems occur with the selection of option.

With kind regards,

Samer

Hi Rafael

Did you know that the Gene Kelly clip should really be called - Singin’ in the Milk.

Plain water didn’t show up on the film so they added milk to it.

The actual song was first performed back in 1926.

Best regards

Mike Testro

MILK TO THE OTHER SOFTWARE AS IN SPIDER PROJECT YOU GET TRUE FLOAT
AFTER RESOURCE LEVELING.

If you still do not see the resource dependences you can add some milk by selecting Create Dependences. Thes will be shown in the Gantt diagram as dotted lines. Once again, remember these are dependences that can change as the schedule is updated and time demand for resources changes. These are not fixed and in any way will interfere with true resource leveling.



Best regards,
Rafael

Dear Rafael,

I fully agree on the resource dependencies. But I still do not agree that CPM scheduling changes from one software to another. The basics are the same, what changes you make to the choices in the input will make changes in the output.

With kind regards,

Samer

Samer,

For a start different software use different resource leveling algorithms, even in Spider Project you have the option of several algorithms, a truth not to shame not to hide, on the contrary to be openly exposed.

There is also another difference in computation of Early Start where this value is dragged to the right in most PDM computations but that can be modeled to be at the left under PDM computations, this is solved by P3 when you allow for activity splitting. I believe it can be better modeled under PDM without activity splitting and just to specify if drag to left or drag to right. The remaining is mathematical computations based on the different types of float. I would like it to be available not at the project level but at the activity level.

On the issue on continuous/discontinuous (non-split/split activities) the following article can serve as a start.

http://www.pmicos.org/topics/aug2006tom.pdf

As Vladimir once said AOA and PDM are just different representations of the same.

Best regards,
Rafael

Dear Samer,
CPM schedules will be the same if the original data are the same and the software the same way interpret these data.
For an example, ALAP activity means different for MS Project and P6 (and Spider).
But everything changes if resources are restricted. CPM does not work anymore and the levelling results that are produced by different software are not the same.
We are proud thatSpider Project calculates the best (shortest) resource constrained schedules.
Please inform me about the software that you use and I will send you the examples.
Best Regards,
Vladimir

Dear Vladimir and Rafael,

Thank you for your kind explanation. I am using P6.7.

Why do you want to use resource leveling? Of course the software will produce outputs depending on its programming. And that is a black box only the mother company knows.

Keep it simple and control the Schedule. Let the computer do the number crunshing for the forward and backward pass. And with a systematic trial and error, you can achieve the contractual milestones, by changing the planned activities, resources and duration.

For complex project, follow systematic procedures, and tightly monitor the input and output.

The core of CPM is intact and is the same in all softwares. The add on is what is causing numerous degrees of difficulties for people to understand, use, and explain the results. And I am very much against that. The Schedule Model should be like any other system in Engineering. Unless you can predict exactly how the model is going to behave, then it is rejected. People are submitting programs of works and they can not explain the output. This is totally unacceptable.

I think maybe in the future I should get a copy of Spider and compare results. Maybe by beginning of next year or earlier.

With kind regards,

Samer

Samer,

Black Box is not knowing true float after resource leveling, even on a two activities schedule Primavera (all versions), Microsoft and Asta gives you the wrong answer.

Proposing not using resource leveling functionality on thousands of activities is kind of insane.

I agree with you with regard to any software that give you wrong float after resource leveling, better do it at the 1950’s way, by hand as the 1960’s is still wrong, and worst, is misleading.

Best regards,
Rafael

Dear Samer,
if resources are limited, if there are financial restrictions, if logistics set restrictions on supplies and the schedule does not take this into account the schedule will be unrealistic.
The purpose of creating schedule model is to get help with decision making. If project manager can get the answers to the questions like what if we will use additional resources, what if we will get the loan, etc. using computer model then it is worth the efforts for its creation.
Resources are always limited. The client may ignore this but not the contractor.
Best Regards,
Vladimir

Dear Rafael and Vladimir,

I agree that the Schedule is very important for the "what if" analysis, and it is used for many decision making actions pertaining to recourses. Actually, this can not be done in a timely manner without these tools.

I am very much against trusting the black box without understanding its formula. The problem is that people start applying the constraints and lead and lags, calenders and local and global options without understanding the black box formula. They you have an output that is support to predict the time and cost throughout the next couple of years.

Unless you have a systematic process to verify the input, and make sure that you Model is correct, there is no way to make certain that the output is correct unless you rely on experienced engineers with prior experience in a similar project.

With kind regards,

Samer

Hi Samer

I have had the same knockabout chat with these two eggheads on the topic of resource levelling on a number of occasions on PP.

I am with you Samer - I don’t trust any computer to make my decisions - especially when it won’t tell me what it has done to my schedule.

I now let Raf & Vlad speak to each other on their own level where they can do least harm to the rest of us.

Best regards

Mike Testro.

Hi Mike,
we live in a free world and everybody can do what he/she likes if it does not harm anybody.

I can imagine that in your projects resources are unlimited, Unfortunately in most projects where we are involved there are resource restrictions and creating project schedules we cannot ignore them. Our schedules consist of many thousands activities and manual levelling I consider as a joke.

Most people just draw their schedules and their schedules do not answer to what if questions. They are almost useless. When people manage projects they manage resources and if the model does not answer to the questions on resources it does not help much.

Best Regards,
Vladimir

Dear Mike,

Many thanks for your kind support. I highly appreciate it.
I think that you should have entered this discussion a while back, but it is not too late.

Dear All,

For me a Scheduling Programs are a big calculator. Everyone can use the +, -, *, /, = and punch in the numbers and get the right answer. Unfortunately, not everyone can juggle few thousand activities using windows and then VERIFY the right answer.

The level of mathematics envolved in programming the box is high, and an engineer (Junior or Senior) does not have the time to understand it (even if can understand it, his company is not paying him to understand it), bearing in mind that, it has to be understoond and verified by the other party. So, it is better to assume that the black box is giving the right answers until it is critical enough project to veryify the results using another black box.

History speaks for itself. Civilizations were built without computer or programs. CPM came in the 1950’s. Needless to say, that current civilizations underwent TWO world wars without computers or programs.

I can very much assure you ALL, that Engineers CAN build construction projects successfully without ANY computers or ANY Schedules. If the Romans, Egyptians, and Chinese succeeded in building Empires for hundreds and thousands of years, then my statement is valid. Can you tell me the name of the software that enabled the British Empire (largest Empire in history) to rule for hundreds of years! The answer is No. You agree with me that it required a lot of planning and scheduling. That is why, I am confident that today’s Engineers can do the same thing if they are put under similar circumistances, for the simple reason that they have more education and more understanding of Engineering than the "Engineers without Degrees" hundreds of years ago.

Let me state very clearly that the CORE purpose of a CPM software is to SAVE TIME AND MONEY. If the softwares that engineers are using these days are not providing this, then the original purpose is lost. These tools are adding more confusion and are not being used properly. Gates was successful in creating a "friendly window environment" that other are using as "input" to the black box, and "output" to view the results. Scheduling Software are like tools in the hands of users. Can a T-square draw a perfect drawing! You are mistaken to think it can. Even if it has a buzzing name.

Ultimately, scheduling tools must evolve in order to enable the user with "reasonable" experience to use the tool to save him/her time and money, and for this result to be verifed (i.e, a checking mechanism).

With kind regards,

Samer

Samer,

I tried to illustrate the issue on how wrong and misleading many software are by not being able to get it right within a simple two activities schedule, I don’t believe a computer is necessary to understand the five activities schedule either. Is not a black box, is an easy to verify fact.

That different software can yield different results and maybe only one can provides you with true float after resource leveling is no reason to throw to the floor these tools, they can still be of some use.

Do not throw to the floor your CPM software as if no longer wanted toys, even if not high end, even if used as a low end calculator or an abacus.

Best regards,
Rafael

Dear Rafael,

Thank you for your kind reply. I think that the softwares that we have nowadays are doing a great job at solving problems. And I hope that their black boxes are programmed correctly.

I saw that you referred to the abacus earlier and that it is not designed to solve the decimal points. So I started a small google search and found the following the you might find interesting about the accuracy of "human" before the word computer came to existance:

http://en.wikipedia.org/wiki/Abacus
" Native American abaci
.....
Altogether, there are 13 rows with 7 beads in each one, which makes up 91 beads in each Nepohualtzintzin. This is a basic number to understand the close relation conceived between the exact accounts and the natural phenomena. This is so that one Nepohualtzintzin (91) represents the number of days that a season of the year lasts, two Nepohualtzitzin (182) is the number of days of the corn’s cycle, from its sowing to its harvest, three Nepohualtzintzin (273) is the number of days of a baby’s gestation, and four Nepohualtzintzin (364) complete a cycle and approximate a year (1 1/4 days short). It is worth to mention that in the Nepohualtzintzin, amounts in the rank from 10 to the 18 can be calculated, with floating point, which allows calculating stellar as well as infinitesimal amounts with absolute precision."

Now this is the first time that I read that article, but I found it interesting that they are using "floating" and "absolute precision" for a systm that was invented by humans 4000 years ago.

If you dig a little deeper and click on the Mayan culture and CONSTRUCTION, you will find the following:

"The Classic period (c. 250–900 AD) witnessed the peak of large-scale construction and urbanism, the recording of monumental inscriptions, and a period of significant intellectual and artistic development, particularly in the southern lowland regions.[9] They developed an agriculturally intensive, city-centered empire consisting of numerous independent city-states. This includes the well-known cities of Tikal, Palenque, Copán and Calakmul, but also the lesser known Dos Pilas, Uaxactun, Altun Ha, and Bonampak, among others. The Early Classic settlement distribution in the northern Maya lowlands is not as clearly known as the southern zone, but does include a number of population centers, such as Oxkintok, Chunchucmil, and the early occupation of Uxmal."

You will agree with me that these guys 4000 years ago, planned well and scheduled their Construction sites well. They had limited resources and were able to build structures with local materials that lasted until our present day.

With kind regards,

Samer

Samer,

About the Abacus thanks for highliting about the decimal point is amazing. And they only used base 20, imagine if using base 21 (boys edition abacus).

Best regards,
Rafael

Hi All

The standard abacus is a marvelous calculating machine - I never mastered it although I tried.

I once worked with a Japanese company in Baghdad and the Project manager did all his calculations on an abacus.

The European engineers were intrigued because thay used somewhat rudimentary calculators from Texas Instruments. (this was in the late 80’s pre computers)

I set up a challenge of abacus v calculator with a somewhat complex calculation - something like:

((197*42)/18)+(14.57*28.6)=??

The abacus won by a fair margin to 2 decimal places - 876.37.

Best regards

Mike Testro

Dear Rafael,

The base 21 idea gave me a good laugh. Thank you.

With the help of google and few ideas looking at a schedule in front of me, I decided to find out the history of paper, mathematics, electricity, computing tools, CPM, and here are the findings:

Paper:
1. Early people discovered that they could make simple drawings on the walls of caves, which was a great place for recording thoughts, but wasn’t portable.
2. About 5,000 years ago, Egyptians created "sheets" of papyrus by harvesting, peeling and slicing the plant into strips
3. 3,000 years. The person credited with inventing paper is a Chinese man named Ts’ai Lun.
4. 10th century, Arabians were substituting linen fibers for wood and bamboo, creating a finer sheet of paper
5. In 1448, Johannes Gutenberg, a German, was credited with inventing the printing press


Tools
1. The original compact calculator was the abacus, developed in China in the ninth century.
2. The young French mathematician Blaise Pascal (1623-1662) invented the first adding machine in 1642, a clever device driven by gears and capable of performing mechanical addition and subtraction.
3. The first commercially successful adding machine was developed in 1886 by William Seward Burroughs (1855-1898).
4. The "Millionaire," a machine invented by Otto Steiger in 1894, was the first adding machine also capable of direct multiplication.
5. The Atanasoff–Berry Computer (ABC) was the first electronic digital computing device.[1] Conceived in 1937, the machine was not programmable, being designed only to solve systems of linear equations.
6. The hand-held pocket calculator was invented at Texas Instruments, Incorporated (TI) in 1966 by a development team which included Jerry D. Merryman, James H. Van Tassel and Jack St. Clair Kilby

The numbers:
1. The Arabic numerals or Hindu numerals or Hindu-Arabic numerals are the ten digits (0, 1, 2, 3, 4, 5, 6, 7, 8, 9). The decimal Hindu-Arabic numeral system was invented in India around 500 AD.

2. While the word algebra comes from the Arabic language (al-jabr, الجبر literally, restoration) and much of its methods from Arabic/Islamic mathematics, its roots can be traced to earlier traditions, most notably ancient Indian mathematics

2. Prehistoric mathematics: The origins of mathematical thought lie in the concepts of number, magnitude, and form.[8] Modern studies of animal cognition have shown that these concepts are not unique to humans. Such concepts would have been part of everyday life in hunter-gatherer societies. That the concept of number evolved gradually over time is evident in that some languages today preserve the distinction between "one", "two", and "many", but not of numbers larger than two.[8]
The oldest known mathematical object is the Lebombo bone, discovered in the Lebombo mountains of Swaziland and dated to approximately 35,000 BC.[9] It consists of 29 distinct notches deliberately cut into a baboon’s fibula.[10] There is evidence that women used counting to keep track of their menstrual cycles; 28 to 30 scratches on bone or stone, followed by a distinctive marker.[11] Also prehistoric artifacts discovered in Africa and France, dated between 35,000 and 20,000 years old,[12] suggest early attempts to quantify time.[13]
The Ishango bone, found near the headwaters of the Nile river (northeastern Congo), may be as much as 20,000 years old and consists of a series of tally marks carved in three columns running the length of the bone. Common interpretations are that the Ishango bone shows either the earliest known demonstration of sequences of prime numbers[10] or a six month lunar calendar.[14] Predynastic Egyptians of the 5th millennium BC pictorially represented geometric designs. It has been claimed that megalithic monuments in England and Scotland, dating from the 3rd millennium BC, incorporate geometric ideas such as circles, ellipses, and Pythagorean triples in their design.[15]

4. The most ancient mathematical texts available are Plimpton 322 (Babylonian mathematics c. 1900 BC),[1] the Rhind Mathematical Papyrus (Egyptian mathematics c. 2000-1800 BC)[2] and the Moscow Mathematical Papyrus (Egyptian mathematics c. 1890 BC). All of these texts concern the so-called Pythagorean theorem, which seems to be the most ancient and widespread mathematical development after basic arithmetic and geometry.


Electricity
1. Benjamin Franklin and Electricity In June of 1752, he performed his famous kite experiment, drawing down electricity from the clouds and charging a Leyden jar from the key at the end of the string.
2. Thomas Edison’s greatest challenge was the development of a practical incandescent, electric light. Contrary to popular belief, he didn’t "invent" the lightbulb, but rather he improved upon a 50-year-old idea. In 1879, using lower current electricity, a small carbonized filament, and an improved vacuum inside the globe, he was able to produce a reliable, long-lasting source of light.

CPM

CPM was the discovery of M.R.Walker of E.I.Du Pont de Nemours & Co. and J.E.Kelly of Remington Rand, circa 1957. The computation was designed for the UNIVAC-I computer. The first test was made in 1958, when CPM was applied to the construction of a new chemical plant. In March 1959, the method was applied to a maintenance shut-down at the Du Pont works in Louisville, Kentucky. Unproductive time was reduced from 125 to 93 hours

Now we are using all of the above to save time and money on our construction projects. Now I have to say that my favorite was discovering tha the oldest known mathematical object is the "Lebombo bone".

With kind regards,

Samer

Hi Samer

A excellent piece of research but you missed out some English contributions:

1. Charles Babbage - Difference Engine - 1849 - a fully programmable calculating machine.

2. Bletchly Park WW11 code breaking computer constructed by Alan Turin circa 1941

3. Clive Sinclair first pocket calculator using LED’s early 1960’s

And it was the Ancient Greeks who discovered electricity by rubbing an amber rod with silk to generate powerful static charges - the Greek word for amber is electron.

Best regards

Mike Testro

Dear Mike,

hahah. Thank you for the remark. I will do better next time.

With kind regards,

Samer

Hi Samer

I forgot the most important one.

Circa 1650 the invention of the slide rule following John Napiers discovey of logarithms.

Best regards

Mike Testro

Dear Mike,

Yes the Slide Rule is very important discovery. It lasted 300+ years.

But I still think the newest information is the "Lebombo bone" from 35,000B.C.

With kind regards,

SAmer

Dear Rafael,

I came accross a very interesting article published in the ENR magazine in 2003:

" Critics Can’t Find the Logic in Many of Today’s CPM Schedules." It is a 4 page article quoting "Priests" and "tool makers" in this field.

www.cpmguru.com/.../ENR%20-%20Critics%20of%20Todays%20CPM%20Schedules.pdf

So, we can reasonable say that up to 2003, according to published sources, that the state of Scheduling is a mess.

The basic need of the Owner/Construction Manager to save time and money is not met "easily" by the available tools in the market. The user-tool interface is a mess. Users can’t transform their ideas into CORRECT networks of logic easily, in order for the black box to calculate it correctly.

I still think that we need to get the guys in a brainstorming session and transform that "Lebombo bone" into a modern tool. Think of it like an ipod. On one side you input the data and on the other side you get the time and cost savings.

I think that I just came up with the next ORACLE market penetration idea :) load your software of choice on ur ipod for $9.99/mo :)

With kind regards,

Samer

Hi Samer,
what is your proposal?
Remember track "Let’s Challenge Spider"?
Ask or suggest something that is required but is not supported.

Spider Project is a tool for finding proper management decisions based on the information that was entered. If you created reliable schedule model then the software shall provide you with the optimal or near optimal solutions for the project schedule, resource assignments and usage, that take into consideration all existing constraints.
What is missing?

The paper that you mentioned described the oppinions of some people. By the way none of them knew Spider Project. Do you agree with them?

Best Regards,
Vladimir

Dear Vladimir,

First of all, let me tell you that I hope that I have the opportunity to work on your software in the future.

The primary issue that I have is that a tool must serve the Owner/ Construction Manager in taking the right decision on the project to save time and cost. I also believe that this is much needed at any point in history.

I can safely say that according to my experience a project of a budget of around ($5 Million, 2 years) can be completed with without anytools whatsoever. The only scheduling that is needed is an excel sheet and a calculator. The Construction Manager can draw the schedule with excel on ONE page and monitor it and reduce the cost on a daily basis. I have done this at least 50 times.

On larger projects, where the uncertainities are more, different people are working on different tasks, the issue becomes harder. The actual plan is difficult to FORM. Time is needed to create it, and many other factors come into play. For the sake of argument let us assume that the only issue here is to come up with a plan to save time and money. Speaking from my experience, I get 5000 activites (150 pages) projects frequently, and I review and return them within 4 hours for the simple reason that it is not representing the project. Most frequently, a planner prepares the Schedule, and the experienced people do not have the time to check the Method Statement, Constructabilities, Deliverables, relationships, and Contractural constraints.

I am confident that you program Spider is a magnificent tool. The issue in not with the black box. The issue is with creating the model. The Construction Manager/ Owner (who are the decision makers) must be able to communicate directly with the software to take the important decision. Nowadays, the tools are driven by Schedulers who know how to use the tool (to some extent/ depending on their experience) without the proper interaction from management. This tool is used to create Schedules to show contactual milestones, and unless the other party is experienced enough to debug the submitted program, it become a contractural document that is supposed to be monitored and tracked to achieve the project.

My suggestion is simple, Model Creation must be simple enough like punching the buttons on a calculator. I think that you need to communicate with Texas Instrument and create such a tool.

I also suggest that you send the 4 guys copies of Spider with all the supporting documents. Let them test it and use it and next they sit down and have coffee, they would be able to talk about something of value available on the market and ENR can publish the state of the available tools around the world. Of course, I understand that there are many other issues at play here, but it seems that they guys genuinely want to improve the status.

No body is talking about the Black Box, everyone is assuming it is working correctly. The issue here is that getting the right information, in the right order to create the right model is something left for specialist.

With kind regards,

Samer

Dear Vladimir,

First of all, let me tell you that I hope that I have the opportunity to work on your software in the future.

The primary issue that I have is that a tool must serve the Owner/ Construction Manager in taking the right decision on the project to save time and cost. I also believe that this is much needed at any point in history.

I can safely say that according to my experience a project of a budget of around ($5 Million, 2 years) can be completed with without anytools whatsoever. The only scheduling that is needed is an excel sheet and a calculator. The Construction Manager can draw the schedule with excel on ONE page and monitor it and reduce the cost on a daily basis. I have done this at least 50 times.

On larger projects, where the uncertainities are more, different people are working on different tasks, the issue becomes harder. The actual plan is difficult to FORM. Time is needed to create it, and many other factors come into play. For the sake of argument let us assume that the only issue here is to come up with a plan to save time and money. Speaking from my experience, I get 5000 activites (150 pages) projects frequently, and I review and return them within 4 hours for the simple reason that it is not representing the project. Most frequently, a planner prepares the Schedule, and the experienced people do not have the time to check the Method Statement, Constructabilities, Deliverables, relationships, and Contractural constraints.

I am confident that you program Spider is a magnificent tool. The issue in not with the black box. The issue is with creating the model. The Construction Manager/ Owner (who are the decision makers) must be able to communicate directly with the software to take the important decision. Nowadays, the tools are driven by Schedulers who know how to use the tool (to some extent/ depending on their experience) without the proper interaction from management. This tool is used to create Schedules to show contactual milestones, and unless the other party is experienced enough to debug the submitted program, it become a contractural document that is supposed to be monitored and tracked to achieve the project.

My suggestion is simple, Model Creation must be simple enough like punching the buttons on a calculator. I think that you need to communicate with Texas Instrument and create such a tool.

I also suggest that you send the 4 guys copies of Spider with all the supporting documents. Let them test it and use it and next they sit down and have coffee, they would be able to talk about something of value available on the market and ENR can publish the state of the available tools around the world. Of course, I understand that there are many other issues at play here, but it seems that they guys genuinely want to improve the status.

No body is talking about the Black Box, everyone is assuming it is working correctly. The issue here is that getting the right information, in the right order to create the right model is something left for specialist.

With kind regards,

Samer

Samer,

The paper is related to a complaint about the status quo of PDM software at that time, yes they were correct. At that time and for decades software developers did nothing practical to the resolution of this problem same as Primavera who limited to buy P3 from another developer, then SureTrak also from another developer, then P3e also from another developer that eventually came P6 because of the many bugs and finally to end up being acquired by Oracle. Other software developers are in the same boat, adding nothing relevant but cosmetic.

Yes PDM is more complex than the CPM of the 60’s because it expands the modeling possibilities, probably it was too much for these developers that never introduced much needed functionalities to understand it. Unbelievable, but initially CPM did not take into considerations for limited resources. Take for example resource float, without float you only got a Bar Chart, after resource constraining how can you understand your CPM without true resource float? How can your contract talk about float if the required software is incapable of providing you with the true value of this metric?

About the complaint on how some other functionalities such as constraints that mess with accepted CPM computations I agree, if the software developer insist on these being available then just avoid them in your modeling. Unfortunately these developers never provided for this metrics to be available for comparison purposes without an easy way to toggle them on and off forcing some of us as well as many analysts to either remove them temporarily or keep a second model without the constraints that fool the PDM backward computations. This is an issue when it is the Owner who requires you to include such implementation of constraints.

As far as I recall in this paper there was also the issue about the technical competence of those doing CPM scheduling, this is not for everyone, the scheduler must be knowledgeable of what is being modeled and must be capable to deal with the available software at the time. Don’t blame the tool for its irresponsible use, start requiring that the persons in charge of the creation of the model be capable and competent; this includes the scheduler as well as the project manager, if the project manager is incapable of dealing with CPM at least require for a capable assistant.

Spider Project team is making all attempts to make it possible for you to understand the logic, they provided us with true resource float and are moving forward with other functionalities with this purpose, issues other developers don’t dare to explore. The concepts are not difficult to understand, the programming is, we should encourage the developer to tame the software for us. Spider team is raising the bar and this will eventually make others to follow for the benefit of all, in a way PP is part of this, we represent the end user, wise developers are listening.

Best regards,
Rafael

Dear Vladimir,

Please read the opinions of other as well on the same issue. Amazingly it is the same:

pscinc.homestead.com/CPM.html

With kind regards,

Samer

Dear Samer, First of all, let me tell that I agree with you. More than that I am strongly against project schedulers that draw the schedules. When implementing PM System we suggest to create a set of databases (reference-books) including, Resource Dictionary, Material Dictionary, Cost Components Dictionary, Material Requirements per unit of volume for typical activities and assignments, unit costs for typical activities and assignments, Typical Resource Crews, Resource Productivities for typical assignments, etc. The usage of these databases is mandatory. Besides, PMO creates a set of typical fragments - small projects that model typical technologies (like construction of 1km of the pipeline, making 30m2 wall, etc.). These fragments are created with the use of above databases. Creating new project model is very simple. It is enough to create WBS or use the template and then use Replace the phase by the new project function selecting requirent fragment and adjusting the volumes of work. With the corporate reference-books and libraries it is easy to create new project models, these models are more reliable (are based on the corporate norms), everybody understands what data are in the foundation. This is the right way to create schedule models.

Dear Rafael,

I did not think that the guys in the 60’s messed up. They had 3000 projects running at the same time and they wanted to land on the moon. So I think that it is safe to say that if you want to build a project on earth for $5M or less, you do not need a progam. You need a calculator only. On the other hand, if you want to go outside the Earth, they you need computers for communication back and forth. I think that construction (like LEGO) is done manually up their as well.

On the other hand, if you have layers and layers of people all over the place, and you are paying them. Then chances are if you buy the black box, and create a model, some kind of calculations is better than nothing. A little order is good. And maybe, just maybe, if you hire the best guys to create the model and maintain the system, then you can get COST AND TIME savings.

With kind regards,

Samer

Dear Rafael,

I did not think that the guys in the 60’s messed up. They had 3000 projects running at the same time and they wanted to land on the moon. So I think that it is safe to say that if you want to build a project on earth for $5M or less, you do not need a progam. You need a calculator only. On the other hand, if you want to go outside the Earth, they you need computers for communication back and forth. I think that construction (like LEGO) is done manually up their as well.

On the other hand, if you have layers and layers of people all over the place, and you are paying them. Then chances are if you buy the black box, and create a model, some kind of calculations is better than nothing. A little order is good. And maybe, just maybe, if you hire the best guys to create the model and maintain the system, then you can get COST AND TIME savings.

With kind regards,

Samer

Dear Vladimir,

We are talking about the same thing from different angles. You are the mathematician and I am the site person.

I am sure that in order to complete the requirements in your 1st paragraph, you need a capable system administrator to do the job with a great setup. It would be interesting to find out how many sites on earth have these capacilities. My guess that you need a project between $15-50Million, to start thinking about this setup. Roughly speaking you will need to spend 1-2% of the project value to start speaking this language.

Now this is the important question. A Program is a normal requirement on the project, from your experience, paying the 2% are you getting your ROI or are you wasting money and getting in trouble because of bad models.

With kind regards,

Samer

Dear Samer,
Spider Project is used by the companies that manage $100000 projects (flats repairs and similar projects) and companies that manage $30bln programs like preparation of Olympic Games (arenas, hotels, railroads, sea ports. airport, utilities, power plant, etc.). I don’t agree that there is some mimimal budget when it is useful. My experience shows that it is always useful.

This company that repairs flats is even more happy because they saw immediate benefits having full information on required materials, resource schedule, cost forecasts, etc.
And the benefits are much larger than 2% of the project budget. They paid $10000 for the software, training and consulting and returned their money immediately - their customers were happy to pay more money for fast, reliable and clean project management when they know the exact days when the work will be done and can track the progress. They also saved money buying necessary materials in necessary quantities at required time.

Our other client built the high-rise building using just in time technology. They called to our office and informed us that the construction was finished at the exact day that was planned one year earlier and we congratulated them. This building was built in the city center and without the precise resource scheduling they may have a lot of problems.

This setup job is necessary, takes time and requires dedicated team. But this job is needed in any case - there is a need in the corporate standards and norms if the company wants to manage its resources and costs. I cannot imagine corporate project management system without corporate standards for production rates, cost and material estimates, etc.
But if these standards are established project scheduling became easy and reliable.
We spent more than 3 months creating norms and resource databases, typical fragments and WBS templates for Caspian Pipeline Construction Project.
After this was done it took three days to create the detailed schedule model, to make all necessary what if calculations, to determine what resources are needed at any point in time, etc.
Our analysis showed that without this model we would be late at the rice fields and the project could be at least half of the year late.
Our next project was Blue Stream pipeline and everything was ready in two weeks because we already had necessary databases and fragments.

And I want to add that I am mathematician but was and am involved in planning and management of many construction and other projects. Creating useful PM software it is necessary to have site experience. When we implement PM System and Spider Project in new company we manage pilot project together sharing with our customers our practical experience. And it includes not only software implementation but also project management procedures and decision making, staff motivation, communications, etc. The software is a part of the project management system and does not work by itself.

We estimate the benefits of proper planning and management in at least 10% of the project budget. And if PM system implementation will cost less then the company will get its money back on the first project. And usually it happens this way.

But I mean the implementation of the project management system and resource scheduling that is based on real data. Drawing schedules does not save much money.

Best Regards,
Vladimir

Dear Vladimir,

This is very interesting information. I think that if you can make a presentation and put it on spider website, it would be useful for many.

The issue of training a staff from zero, building a database and make them productive seems ideal. But I will take your word. It can happen, but that is not usually the case.

What you are describing is a project management solution with providing services to the Contractor. Chances are that you Schedulers are building the Model for the Contractors. That is why it is working. In most cases, the Contractor buys the software, and hires a scheduler to do the job.

It seems to me that the solution is in running the black box correctly. We are talking about creating an Interface between the site and the black boxes. This consist of prefessional service companies that can run the software and understand construction at the same time. This is a partial solution because the information at site changes by the minute. And when you have 50 people working 1 or 2 shifts, things really change daily. In addition, the majority of the users will not allow a 3rd party to have any part of their data!

With kind regards,

Samer

We don’t advise our customers to use external services.
If construction managers did not take part in the schedule development the schedule is good for wall paper.
The contractor schedule shall advise site managers what works shall be done in the next shift, procurement managers shall have the information when and what amount of materials will be required, etc.
Contractor project schedule shall supply project participants with the information that is required for his/her tasks. It shall be updated frequently and deal with physical work volumes, real costs, real resources.
It shall be created inside the organization and shall be based on the internal corporate norms and technologies.
It is necessary to train contractor managers because schedule development lasts until the project finish.

The practice that you described (hiring external scheduler) is not common in the countries where we work. And the task of PM System implementation is to teach our customers to manage their projects without external help.

Best Regards,
Vladimir

Dear Vladimir,

The CPM was created at a Business oriented company to serve a certain need. It was later determined that this tool did not work by the same company, and the founders had to leave and open their own practices. It was further devleped at research institutes and huge organizations.

It is very obvious that the origin of CPM did not start with the end customer in mind. It is a systematic approach managed by specialist. What we have in the market are intelligent tools that needs "Certified" people to run it.

What I am talking about is something different. Construction Projects are initiated by Owners, Managed by Project/ Construction Managers and executed by supervisors. These three categories do not have tools to reduce the project time and cost within their immediate reach except a calculator and to the extent that any of them is computer literate.

What are the avilable solutions to the time and cost decision making saving beside using their experience: Owners drink and smoke to relax their nervous systems, PM/CM look at the skies to receive wireless clues and supervisors have calculator and drawings in their hands running around to complete the job as efficiently as possible and if the PM/CM is smart, then it will effective as well.

What do you expect of an Owner/PM/CM if you give him a monthly report of 100 Page CPM with pictures and cummulative graphs, etc. How do you expect them to take the right time and cost savings. Especially since every single day at site labor cost, is more expensive than the tool.

The more important question is that how do you expect people who are hired on a monthly basis by the Owner to provide him with solutions to complete the job faster. It is not possible logically because it is not covered contractually in most cases, and it means that your are asking people to reduce their income.

If you can explain to me logically how your average Owner who does not know anything about CPM can save time and cost by using the available CPM tools in the market, I would very much appreciate that. They are the entity most interested in time and cost, and they can’t use the tool. And I am very confident that they know everything about money and interests.

The current tools, to my knowledge, are managed by specialist, and are good for progress monitoring. You can have a indication about what MIGHT be happening at site depending on how good the model it.

We are talking about a different tool. That is why the oldest schedule in history marked on a bone, can be a good start. If it was easy to be understood by a 35,000BC person, any living person now can use it.

With kind regards,

Samer

Dear Samer,
if project management does not use created schedules then they are useless.
Construction manager shall be involved in the schedule development and will require from supervisors to follow the planned order of activities execution. The schedule defines how many people shall work at any site area at any type of job. Out of sequence execution means penalties to supervisor.
Crews get their work plans directly from the scheduling software and these plan shall include planned physical amounts of work to be done.
Reports go to the scheduling program and the schedule is recalculated. The work plans for the next shift (day) are published.
It is usual to have computer (notebook) on site. And it is usual to have phones on site. So it is not hard to connect to the Central office, to submit the reports and to get the revised plan if you don’t want to train supervisors.
Another option that we use here on some projects - the supervisor comes to work 20-30 minutes earlier, switch on the computer, updates the schedule and prints work plans for construction crews.
At the end of the shift crew leaders fill (on paper) the fields against each planned activity with actual quantities that were done by their crews. These reports are submitted to the supervisor who may accept or rejext them but his acceptance means quantities confirmation.
The supervisor enters actual data in the notebook and send to the office or in the scheduling program creating new updated project version.
Information from the construction site, from supplies department, from financial department is merged in the PMO.
Reports to the owner are made by PMO and we suggest to show project trends, the best - success probability trends if risk simulation was approved. 100 pages report will not be read.

In any case - if the schedule lives by itself without guiding production force it is useless.