Playlist
In this course, we explore how to analyze historical data to forecast future outcomes, setting the stage for the utilization of machine learning techniques. While not a prerequisite for machine learning application, this skill set empowers you to critically assess machine learning results and infuse your own analytical nuances. No extensive mathematical knowledge is required—just a grasp of averages and standard deviations, which will be thoroughly explained.
In this tutorial, the initial step involves downloading project data. The focus then shifts to isolating and analyzing specific areas, such as construction progress. The construction activities are easily accessed and downloaded for further analysis.
In this tutorial, we delve into the technical intricacies of project data analysis. Upon downloading the data, the formatted table offers a comprehensive view, encompassing critical details like progress activity, WBS matching, start and end differences, duration, and baseline information. The emphasis is placed on precision, with a calculated average delay for activities in the construction phase, providing nuanced insights into project performance. The tutorial also addresses the identification and handling of invalid dates, primarily attributed to scope creep, setting the stage for deeper exploration in subsequent videos.
00:01 So when you download the data, this is the kind of table you see. Now I’ve formatted it for ease of use. You get progress activity ID, name, whether or not the WBS matches between the two versions – this lets us know scope creep, current start and start difference, current finish, finish difference, duration, total flow, baseline start and baseline end. And what that enables us to do is very quickly take and understand the start difference, which is simply just read from the table, and the end difference, which is simply read from
00:39 the table. And what I’ve done to keep things very simple is I’ve calculated the duration difference. Now obviously if you have multiple calendars this is a little more tricky but I’ve kept it simple right now. Now what I can see in this, the start and end difference, if I go average of everything in this entire column here, make that smaller, we can see that activities are on average delayed by 78 days starting, they’re delayed by 79 days finishing, and on average they take an extra 0.7 days longer
01:25 to finish in this schedule right here within construction. Now what we can also do is filter on progress and see just the activities that are completed, and that is where we’re deriving this data from. So we can see this data set here. Now these invalid dates here, these are normally driven by scope. So if we take out all of the ones that are only current, you’ll see that all of the invalid dates – so this is scope creep which I’ll come onto in the next video – but very quickly with this kind of technique I can start to hold
02:06 people to account on their delivery performance.
In this advanced analysis, we shift our focus to evaluating the impact of scope change on project dynamics. Utilizing a straightforward counting technique, we determine the percentage change in the number of activities between the baseline and updated plan. By considering both the activity count and weighted duration, we gain a nuanced understanding of the potential scope change impact. The tutorial introduces a weighted approach based on activity duration, providing a more sophisticated assessment. This knowledge empowers project planners to anticipate and strategically manage scope changes, ensuring effective project control and resource allocation.
00:01 Now we’ve analyzed for start, end and duration difference, we can analyze for scope change. And there’s a very simple technique that I like to do here and it’s counting. So what I’ve done is I’ve done a raw count of all the activities that were in the baseline and there’s 46 here. And I’ve counted how many new activities there are and it’s 9, which gives a total of 55. If you’ve done the reviewing and update course before, this is the 55 you see there. Now just by putting a percentage against it,
00:32 so 9 divided by 55, I can see there’s been 16 percent change to the number of activities in this plan. Now what that means is that going forward over the same time period, it would be reasonable to hypothesize that there’s going to be another 16 percent change going there. Or if I wanted to take the 9 out, I could inverse it and do, for example, 55 divided by 46 minus 1, which would be 5 divided by 46 minus 1, should be 19 percent difference, rounding up to 20. So I could go forward
01:09 and say I anticipate there’s going to be a 20 percent change in the number of activities increasing in this WBS. Now rather than just doing activity count, I could weight them by duration. Now the average duration of all the activities that were in the baseline was 6.8 days. The average duration of all of the activities that are newly added is 3.6. So I can understand, well if I’m doing it weighted by duration, 3.6 is roughly half of 6.8. Or if we do it this way,
01:40 it is 52 percent. I could take 52 percent of 20 percent here and say if I was doing weighted by duration, I would expect an increase of 10.2 percent of the overall duration of all of the activities summed together. It’s getting quite advanced, but there’s an understanding of the expectation of the impact of scope change. So based on this, I can say quite safely that if the scope change is being driven by us, then it might be wise to have certain buffer periods in the program to absorb
02:20 this level of scope change. If it’s driven by someone else, well that is the decision of the commercial director.
In this tutorial, we’ve executed a concise yet potent Excel-based analysis, employing averages as our primary metric. Despite the appearance of complexity, the underlying methodology is straightforward and accessible with a focused learning effort. This technique provides a robust foundation for project analysis, offering insights that can significantly impact decision-making.
The critical observation is that mastering this method, requiring just an hour of dedicated learning, bestows a unique proficiency in project management. By revisiting and comprehending this approach, you not only unravel the intricacies but also gain a competitive edge. This technical expertise can be a game-changer, distinguishing you as a proficient project analyst with a skill set that’s not commonly mastered in the field.
00:01 So what we’ve done here is a rather simple Excel analysis using averages, but here’s the key thing. I’m sure going through this you’d have thought, well one, this might be quite complicated, number two, is not many people know how to do this. If you re-watch the videos back a couple of times, you’ll see that there’s actually simplicity in the method, and if you learn how to do this, which should take no longer than an hour, you’ll be head and shoulders above all others who can’t do this.
00:31 So what I’d say is this is something crucial that can help you look amazing in your projects. Please take the time to watch it again and learn.
Configure your risk register with ease. Follow simple steps to add, customize, and manage risk events effortlessly. Learn how to seamlessly import your risk register into the schedule. Empower your project management with this quick and easy risk management guide.
00:01
in this course we’re going to learn how to analyze past data to forecast the future this is in a buildup to using machine learning you don’t need to do this in order to use the machine learning but what this does is it gives you the skill set to critique machine learning and to add your own flavor of analysis it is something that the senior planners that I’ve engaged with have all appreciated so sit back you don’t need to know any more mathematics other than averages and standard deviations which I
00:33
will explain and the whole process should take about 15 minutes enjoy so first of all we need to download our data now what I’m going to do is isolate the area that I want to analyze so I want to look at construction So within progress track I click on construction and I’ve got all of the activities here all I need to do is to click on this Cloud download icon in the top right which I’ll do now and it will export all of that table for me to start analyzing so when you download the data this is the kind of table you see now
01:10
I’ve formatted it for ease of use you get progress activity ID name whether or not the WBS matches between the two versions this lets us know scope creep current start and start difference current finish finish difference duration total float Baseline start and Baseline end now what that enables us to do is very quickly take understand the start difference which is simply just read from the table and the end difference which is simply read from the table now what I’ve done to keep things very simple is I’ve
01:48
calculated the duration difference obviously if you have multiple calendars this is a little more tricky but I’ve kept it simple right now now what I can see in this stter the start and end difference if I go equals average of everything in this entire column here and make that smaller we can see the activities are on average delayed by 78 days starting they’re delayed by 79 days finishing and on average they take an extra 7 days long longer to finish in this schedule right here within construction now what we can also do is
02:37
filter on progress and see just the activities that are completed and that is where we’re deriving this data from so we can see this data set here now these invalid dates here these are normally driven by scope so if we take out all of the ones that are only current you’ll see that all of the invalid dates so this is scope creep which I’ll come on to in the next video but very quickly with this kind of technique I can start to hold people to account on their delivery performance now we’ve analyzed for start
03:17
end and duration difference we can analyze for scope change and there’s a very simple technique that I like to do here and it’s counting so what I’ve done is I’ve done a raw count of all the activities that were in the Baseline and there’s 40 6 here and I’ve counted how many new activities there are and it’s nine which gives a total of 55 if you’ve done the uh reviewing and update course before this is the 55 you see there now just by putting a percentage against it so 9 divided 55 I can see there’s been 16%
03:51
change to the number of activities in this plan now what that means is that going forward over the same time period it would be reasonable to hypothesize there’s going to be another 16% change going there or if I wanted to take the nine out I could inverse it and do for example 55 ID 46 minus 1 which would be / 46 minus one be 19% difference well rounded and go up to 20 so I could go forward and say I anticipate there’s going to be a 20% change in the number of activities increasing in this WB yes now rather
04:31
than just doing activity count I could wait them by duration now the average duration of all the activities that were in the Baseline was 6.8 days the average duration of all of the activities that are newly added is 3.6 so I can understand well if I’m doing it weighted by duration 3.6 is roughly half of 6.
04:53
8 or if we do it this way it is 52% I could take 52% of of 20% % here and say if I was doing weighted by duration I would expect an increase of 10 2% of the overall duration of all of the activities summed together it’s getting it’s getting quite Advanced but it’s it’s an understanding of the expectation of the impact of scope change so based on this I can say quite safely that if the scope change is being driven by us then it might be one to have certain buffer periods in the program to absorb this level of scope change if it’s driven by someone else
05:39
well that is the decision of the commercial director so what we’ve done here is a rather simple Excel analysis using averages but here’s the key thing I’m sure going through this you’d have thought well one this might be quite complicated number two is not many people know how to do this if you rewatch The videos back a couple of times you’ll see that there’s actually Simplicity in the method and if you learn how to do this which should take no longer than an hour you’ll be Head and Shoulders above all others who can’t
06:12
do this so what I’d say is this is something crucial that can help you look amazing in your projects please take the time to watch it again and learn
