HC Speaking Assignment Matrix

[johnpoulson]

Maybe there is a software engineer out there somewhere who has already solved this (common?) problem.

In a growing stake it becomes necessary many times a year to create a new HC speaking assignment matrix.

Anyone who had had to make sure that each member of the High Council gets to every ward with a different companion and that each ward get as many different High Council Speakers throughout the year knows the inordinate amount of time it takes an unorganized brain (mine) to sort through this mess.

If someone has a program where I could plug in the number of wards, number of stake speakers (some Stake Presidents add other members of the stake leadership family to the speaking circuit) and number of months and have a schedule magically appear I would be willing to pay almost anything ;-)

After creating my matrix (manually) I print the information in two tables. The first shows a High Counselors perspective (so he can see which ward he will speak at for any given month) and the second one shows the Bishop's perspective, so he can see who his stake speakers will be for any given month.

Any help, ideas or (miraculously) a link to such a tool would be appreciated.

[atticusewig]

I can't remember any specific program
that would do what you are asking, but
is sounds like something that a project
management or timesheet optimization
program might be able to handle.

If not, it shouldn't be too hard for some
of the brainiacs on this list to put
together an open-source project
to do this for you.

- Atticus Ewig

[mkmurray]

I've actually decided to take on your challenge. I will write a simple little application over the next few days and you can see if it suits your needs.

[BlackRG]

Thinking this through, it seems that the matrix creation probably isn't that complicated because it would follow a pattern - if you don't need it to be random. (Of course this is assuming I correctly understand your problem) Example:

4 speakers, 4 wards:

1. Ward1=speaker1 ward2=speaker2 ward3=speaker3 ward4=speaker4
2. Ward1=speaker2 ward2=speaker3 ward3=speaker4 ward4=speaker1
3. Ward1=speaker3 ward2=speaker4 ward3=speaker1 ward4=speaker2
4. ward=1speaker4 ward2=speaker1 ward3=speaker2 ward4=speaker3

etc.

Assuming this is suitable, then it's probably pretty easy for you to come up with the matrix yourself and from there you just keep cycling through it. I'm guessing your real issue here is developing the reports. If that's the case, this is simply solved with an Excel spreadsheet and some lookup functions, assuming you can use Excel or some other spreadsheet with similar functions. To simplify matrix generation, assuming patterns like the above are suitable, it would also be fairly trivial to set it up so that you feed it the speakers/wards for the first speaking appointment and then have it rotate them for you for successive appointments. I can help you throw something together like this if you'd like.

If you'd prefer something standalone though, then it looks like mkmurry has volunteered to help

[russellhltn]

If you go the Excel route, would it be possible to make it OpenOffice compatible so it can run on the Stake computer?

[BlackRG]

I haven't done a lot of this type of work with open office, but looking at it it appears it would support the needed features. If the intended target is OpenOffice, then I'll just use it instead of Excel rather than having to chance a conversion coming out in one piece :P

Are there people other than CaptainStansbury that would need this?

[WelchTC]

I haven't done a lot of this type of work with open office, but looking at it it appears it would support the needed features. If the intended target it OpenOffice, then I'll just use it instead of Excel rather than having to chance a conversion coming out in one piece :P

Are there people other than CaptainStansbury that would need this?

If you do it in OpenOffice but save it as an excel document (within OpenOffice) then you can be fairly confident that both programs would read it fine. At least that has been my experience for 99% of the documents I deal with. There are exceptions so just test it on both.

Tom

[mkmurray]

Thinking this through, it seems that the matrix creation probably isn't that complicated because it would follow a pattern - if you don't need it to be random. (Of course this is assuming I correctly understand your problem) Example:

4 speakers, 4 wards:

1. Ward1=speaker1 ward2=speaker2 ward3=speaker3 ward4=speaker4
2. Ward1=speaker2 ward2=speaker3 ward3=speaker4 ward4=speaker1
3. Ward1=speaker3 ward2=speaker4 ward3=speaker1 ward4=speaker2
4. ward=1speaker4 ward2=speaker1 ward3=speaker2 ward4=speaker3

etc.

Assuming this is suitable, then it's probably pretty easy for you to come up with the matrix yourself and from there you just keep cycling through it. I'm guessing your real issue here is developing the reports. If that's the case, this is simply solved with an Excel spreadsheet and some lookup functions, assuming you can use Excel or some other spreadsheet with similar functions. To simplify matrix generation, assuming patterns like the above are suitable, it would also be fairly trivial to set it up so that you feed it the speakers/wards for the first speaking appointment and then have it rotate them for you for successive appointments. I can help you throw something together like this if you'd like.

If you'd prefer something standalone though, then it looks like mkmurry has volunteered to help

You also assumed 4 months (in addition to 4 speakers and 4 wards).

Let me ask these questions and see how you all think it should be implemented:

How should the pattern be affected when...

• wards > speakers
• speakers > wards
• months > speakers
• speakers > months
• months > wards
• wards > months

[JamesAnderson]

Some other thoughts that might come into play on such a tool. These affect some of the variables. The only part of the equation that is static is the number of high councilors, that is always 12.

1. Many stakes have less than 12 units, in fact, mine has six units. Conversely, there's a stake in Provo that has 15 units right now, there's talk it will be divided at some point in the near future. For less units, some speakers get a bye some months, for more than 12 units in a stake, a stake president may want to add more stake leaders to the matrix to cover all the units in the stake.

2. Some stakes (varies from stake to stake) do not have high council speakers go out in months that have general conference or stake conference in them, that means there might be only eight months a year. If a given stake does schedule high council speakers in months they have stake conference or general conference, then that makes ten months or all twelve months.

3. Some areas are very fast growing and stakes in those areas may find they need to add units frequently, like maybe a couple in a year or or even less at different times, so there will be a need to adjust the number of units to accomodate the changes 'on the fly' so that the matrix can be updated very quickly.

What needs to be done to handle these types of situations where local conditions vary like this?

[BlackRG]

You also assumed 4 months (in addition to 4 speakers and 4 wards).

Let me ask these questions and see how you all think it should be implemented:

How should the pattern be affected when...

• wards > speakers
• speakers > wards
• months > speakers
• speakers > months
• months > wards
• wards > months

Had to assume something to give an example

Much of this is where we're going to need more detail from CaptainStansbury (or others who feel they have a use for this).

If we assume:
1. All wards have stake speakers on the same Sunday.
2. Only one speaker can visit one ward on a given Sunday (this one might be iffy. Impossible for some stakes, but possible for others).

Then I can answer it thus:

• wards > speakers

If there aren't enough speakers, someone has to do without. You insert blanks and rotate them.

• speakers > wards

If there are two many speakers, then either you group them and assume more than one will speak in a Sunday, or you insert blank wards that speakers rotate through knowing that any speaker assigned to a blank ward isn't speaking that Sunday.

• months > speakers

Months aren't relevant to the problem as far as I can tell. When the pattern reaches it's end, it just repeats itself. You use however may iterations you need. If an offset is needed it hit different places in the pattern, that should be fairly easy to allow for.

• speakers > months

Same as above

• months > wards

Same as above

• wards > months

Same as above


It looks like a 2 dimensional problem to me. The months component is just a representation of how many iterations are needed.

Now if we're in a situation here where a speaker can visit more than one ward on a given Sunday, but can only visit certain pairs (like he has enough time to speak at Ward 1 and Ward 2, but not enough time to speak at both Ward1 and Ward3 on the same Sunday) then you may have just stepped beyond the realm of what can easily be accomplished with a spreadsheet.