Autodesk is hosting a webinar on Flow Generative Scheduling called How to Use AI for Fast and Accurate Production Scheduling on Sep 12 @ 11 AM PT!
Come learn how you can generate multiple resource-optimized schedules, evaluate trade-offs, adapt your schedule of choice as project scopes change, and send it to Flow Production Tracking!
If you have a Flow Production Tracking subscription, you have access to Flow Generative Scheduling!
I’m having trouble registering. I’m being forced to register for the chance to win a cup of coffee. What’s the point of having a tick box to opt in if you can’t actually opt out? I don’t even drink coffee
Hi @Halil - If you register for the webinar, you will receive an email with a link to the recording whether you attend the live session or not. We’ve also updated the form so you are not required to “opt in” to the Starbucks giveaway. Thanks!
Does anyone agree with me that a big pitfall is that it uses duration to determine its data?
Duration is a field we cannot change and is a simple calculation between start and due date.
It’s in no way representative (usually) of the actual human resource time spent on the task.
We usually determine that using the bid field or a combination/calculation of bid vs other data.
If I have a comp task that starts on october 1st and is due october 8th it really does not have to mean that the task itself takes 8 days to complete for 1 artist.
It just means that it can start on october 1st as hopefully lighting is rendered by then, and the artist needs to ideally have an approved version by october 8th.
They need to balance their work to make sure they do this within the alotted “artist bid” time.
Which is what we think it should take.
(This is a simple scenario and everyone does this slightly differently.)
I thought the Resource Planner had a good approach where it gives you the ability to tell it how to calculate the time required by making a new field that can be altered.
Hi @Ricardo_Musch! I do agree with you, I think what you’re describing is very typical - artists might double up on tasks, have down time while they wait on reviews or changes from upstream, etc all of which would mean that it isn’t actually the case that “a task with a duration of X requiring Y artist means that task uses 100% of that artist’s time from start to finish.”
We took an intentionally simplistic route with our Flow Production Tracking integration so that we could more quickly get something into people’s hands and hear what people think is missing. We could definitely consider making this more configurable so that you could do something like specify exactly which field is utilized by Flow Generative Scheduling (e.g. something besides duration.)
There was a suggestion in another post (Durations vs Workloads) to potentially leverage Workload for this as well, do you think that would be helpful?
You will be eailed a recording if you signed up for the webinar but it may take a little while as Zoom needs to transcode it etc.
And I’m not sure its automatically sent out.
that may work but I would be carefull to use specificaly that field, since its tied to the Resource Planning app and I’m not sure we would use it the exactly the same way for both Resource Planning and Generative Scheduling (I need to look into it).
But the approach Resource Planning took, giving us control of the data considered for the duration, is a good approach.
We may all do slightly different things to determine the resource needs for a task and what itsactual duration goal is.
I know of studios that use the Bid field but there are also studios that have a Bid field and an Artist Bid field, so they use a calculation to determine the time that stands for a Task.
Others may use a completely custom calculation or calculate the time agains already logged time, there are many options here…
The important thing remains that whatever determines the actual resource requirement or duation for a task, that is what Generative Scheduling should take to determine how many people are needed in that period to keep the schedule from having peaks and valleys.
Flexibility on the data entry point here is key so we can get accurate results out of this.
However basing it “hardcoded” on the duration field makes it instantly not usefull for 90% of the workflows that I am familair with as duration means nothing more than the difference between start and due date of a task, it has no relation to the actual time a human is expected to work on that task, hence the results from Generative scheduling will be skewed and tell me I need x amount of compers when its really not the case.
TLDR
If the duration becomes a controllable setting (like Resource Planning), then I can see this definitely being a very usefull tool and the start of something even greater!
Register here: Webinar: How to Use AI for Fast and Accurate Production Scheduling
