New Member Question (1 Viewer)

daggaz

New member
Local time
Today, 17:17
Joined
Apr 29, 2016
Messages
1
Hi there everyone i am new to using access but would like to learn as much as i can to help me progress my career and my general knowledge i hope you guys can be of help and when my knowledge grows i can hopefully help others in my current situation.

i have an objective or an idea that i want to create but being totally honest i dont know where to start so i will give you guys a bit of info and hopefully you can help me find the relevant section i need to be looking in.

basically at work we use running plans or line up plans i work in a factory on a production line and my zone has 29 stations and we have 34 staff so each day we have certain members of staff on each station but due to holidays or absentees we have to move people around.
i would like to create an automated running plan generator that can assign a worker to a station that they run, not all workers can do every job some can do many jobs some can only do one.
can you guys please advise me as to were i would need to start with this idea and the relevant section to be looking in ?
thanks in advance any help is greatly appreciated
 

plog

Banishment Pending
Local time
Today, 12:17
Joined
May 11, 2011
Messages
11,611
Divide and conquer.

I'd put the big picture goal on the back burner for a bit and break off pieces that will help you learn Access and incrementally get you to the larger goal. What you have proposed is more than simple table layout and queries, its going to require actual programming.

I suggest you take one piece (staff) and build a simple database to manage just that data. Build table(s) to keep track of people (first name, last name, hire date, skills, etc.). Make some reports to get the data out of it, make some forms to edit/input data in it. Then move to another piece (equipment/production line) and do the same. Do that for all the pieces of data you have. Then, bring them together one by one.

Divide and conquer, slowly work toward the end goal.
 

MarkK

bit cruncher
Local time
Today, 10:17
Joined
Mar 17, 2004
Messages
8,178
Welcome to the forum.

Scheduling is a VERY hard problem because there is not ONE solution, there are many, and so you have to do an optimization. You'll need to run scenarios based on weighted values for your many input parameters. How well trained people are on what machines is an input parameter, so you need define minimum training by machine AND machine operator, but these parameters will conflict with who is available based on vacation time, so as you adjust vacation time, you'll need to re-run your optimization for staffing.

So maybe just from this brief description you can see that the problem varies at both ends, and that makes it really hard to validate. Changing staffing changes vacations, and changing vacations changes staffing, and as you explore this problem, I expect you'll find those aren't the only variables.

And then, lets say you write an algorithm that gives you a result. How do you know that result is meaningful or accurate? How do you know it actually provides the best fit? So you'll have to develop some mechanism to assess that "best-fit" apart from your algorithm.

I don't mean to dissuade you, but go to wikipedia and search for "optimization algorithm" and see how vast the topic is.

Hope this helps,
 

The_Doc_Man

Immoderate Moderator
Staff member
Local time
Today, 12:17
Joined
Feb 28, 2001
Messages
26,999
daggaz, I will offer two rules to consider. I call them the "Old Programmer Rules."

1. Access cannot tell you anything you didn't tell it first. Meaning ... if there is something you want out of Access in the form of reports, then you had better be sure that you capture all data needed to support the output.

2. If you can't do it on paper, you can't do it in Access. Meaning ... Access is a rapid application development tool, but it doesn't program itself. You still have to steer even if the wizards and other RAD elements are active. The first step is therefore to see if you can list the various tables and define the data flow between/among them.

In regard to the latter, I often suggest that a viable design method is to get a white-board and a case of sticky-notes. Define the different things you want to track - people (and their abilities/skills), workstations, etc. Use the stickies to represent tables of each such item. Draw lines between table elements to determine how the information interacts to generate whatever it is you wanted to see. If you cannot define the diagrams on the white board, you are going to go nowhere on the keyboard.
 

gemma-the-husky

Super Moderator
Staff member
Local time
Today, 17:17
Joined
Sep 12, 2006
Messages
15,613
as MarkK said, this may be complex. You may research "linear programming" as well.

generally, I would be inclined to start with your scarcest resources first. Allocate the machine with the fewest candidate workers, and/or allocate the worker with the fewest skills. That should get you closer to a workable solution first.

It's like packing algorithms. Generally pack the biggest item first.
 

Users who are viewing this thread

Top Bottom