Excel and Forecasting

[abhalso]

Hello, Can you please send it to me as well. Many thanks
axitawork@hotmail.com

[kgage]

I would like to try out the forecasting tool as well. I have made some in the past, and would be interested to see how you structured yours. Please send me one at:

kory_gage@hotmail.com

Thanks!

[karj]

For a few years I was using MS Excel It could only hold 65,000 records. I finally decided to ugrade and discovered that Excel 2007 now can hold over 1 million records plus many new new features that may be of use to you.

[halycon]

Hello, I would also like to see the I would like the forecasting tool as well, what is the FTP site?

[Chichi]

Dear Tuan,

if u don't mind, could u please sent the software and sthe steps to me too...

nurul_fuad@yahoo.com


Many Thanks

Chichi/Indonesia

[Newguy01]

Hey, I wouldn't mind trying the software....my email address is meltzeredgar@hotmail.com

[pcride]

Is this still available? Do you have a link to download it?

[tophamc]

Originally Posted by thetuan

Hi Adolero,

You can build a very accurate forecast model in Excel but it's not that easy.

Here we go

You will have to create a table where your analyst can input historical information - calls offered, calls answered, ASA and total call processing time (not average call processing time). With this historical information, you can use the pivot tables to build three sets of information. The first pivot table should show Sun - Sat in the column area and the arrival pattern by half hour for each day. This table will give you an accurate look at your daily arrival pattern for each day of the week (in percent). The next table should have your total call processing time by day and by half hour. This table will allow you to forecast the time you will need by half hour to process your work load. The last pivot table will contain information about your daily arrival pattern in a weekly format - this will allow you to forecast your day of week pattern for all seven work days (or the days during the week that your call center is operating). With these sets of information, you can now extract data to accurately forecast your call load curve by half hour and by day. As with any huge amount of data, you will need to smooth out the small bumps within your incoming call load and you will need to normalize all daily volume that is affected negatively or positively by unusual circumstances - for example, Thanksgiving might show a very low volume for Thursday of the Thanksgiving week and you will need to bring that volume back to what is normal for you for Thursday. If you don't normalize your day of week pattern properly, the model will give you an inaccurate DOW (day of week) pattern and your manpower load curve will be off. I can go on about how you will need to incorporate the Erlang theory to forecast manpower load by half hour but that will definitely take us beyond this discussion.

To make all of this easy for you, we have developed a forecast model and scheduling program in MYSQL that will simplify all of this for you. We are looking for folks who are willing to test our forecast and scheduling program and give us feedbacks so we can improve on the model. Let me know and I can give you the link to download the model.

Tuan

Tuan,

Is this still available for testing?

Cece

[thetuan]

All,

I have not checked this forum for a few months now so I am shocked to see the interest in our forecasting and scheduling model. I will post the software on a FTP site later so you can download it and test it out.

Please let us know how this model works for you and give us ideas on how we can improve on the model.

Tuan

[thetuan]

All,

You can download the software at forecast.msi

Once you download and install it, let me know and I can help you to run it.

Tuan

[rohitashhh]

Originally Posted by thetuan

Hi Adolero,

You can build a very accurate forecast model in Excel but it's not that easy.

Here we go

You will have to create a table where your analyst can input historical information - calls offered, calls answered, ASA and total call processing time (not average call processing time). With this historical information, you can use the pivot tables to build three sets of information. The first pivot table should show Sun - Sat in the column area and the arrival pattern by half hour for each day. This table will give you an accurate look at your daily arrival pattern for each day of the week (in percent). The next table should have your total call processing time by day and by half hour. This table will allow you to forecast the time you will need by half hour to process your work load. The last pivot table will contain information about your daily arrival pattern in a weekly format - this will allow you to forecast your day of week pattern for all seven work days (or the days during the week that your call center is operating). With these sets of information, you can now extract data to accurately forecast your call load curve by half hour and by day. As with any huge amount of data, you will need to smooth out the small bumps within your incoming call load and you will need to normalize all daily volume that is affected negatively or positively by unusual circumstances - for example, Thanksgiving might show a very low volume for Thursday of the Thanksgiving week and you will need to bring that volume back to what is normal for you for Thursday. If you don't normalize your day of week pattern properly, the model will give you an inaccurate DOW (day of week) pattern and your manpower load curve will be off. I can go on about how you will need to incorporate the Erlang theory to forecast manpower load by half hour but that will definitely take us beyond this discussion.

To make all of this easy for you, we have developed a forecast model and scheduling program in MYSQL that will simplify all of this for you. We are looking for folks who are willing to test our forecast and scheduling program and give us feedbacks so we can improve on the model. Let me know and I can give you the link to download the model.

Tuan

Hi Tuan,

please send me this link so that i can test it and send you my feedbacks for the same. my email id is rohitashraj@hotmail.com

thanks & regards
Rohitash

[Lawati]

Hi ,

I am willing to try this software as well if that is possible,

allawati786@gmail.com