The Pannet Scheduler README

First off, thank you for downloading our software. This is the first
release of the Pannet Scheduler, and we hope that it proves useful. We
are quite aware, though, that it is likely to be confusing to use and
contain bugs. Thank you for being willing to find out!

In order to install the Pannet Scheduler, you will need the following
files:
  readme.txt   <- this file
  install.doc  <- explains how to install and setup the system
  pandb.zip    <- a blank database. You need it.
  pansched.zip <- the executable files. You need this too.

(.zip files means you'll need an unzip utility. WinZip will work)
-----------------------------------------------------------------

Currently, there is no user manual. We are working on fixing that
glaring omission, and if you'll send us your email address, we'll make
sure you get a copy when we're done (or reasonably close). Until we are
done, here's what we recommend you do. First, explore the sample
database contained in EXAMPLE.GDB (either setup the PAN_SCHD_DB alias to
point to it directly, or copy it to SCHDSOLV.GDB and have the alias
point to that copy (recommended). Once you've got everything setup,
run the scheduler (SCHDGUI.EXE, not SCHDSOLV.EXE!), explore and test.

You should expect to be confused when trying to use the software - the
system is complex and poorly documented. We are working at rectifying
the situation, but feel free to ask questions via email or phone.

In almost every instance, when you change a field on a form, you must
explicity click on a POST button for that information to be entered into
the database. In some instances, as when you are entering information
into a grid-like control (for example, any of the "basic" tables), any
changes will be automatically posted if you move (using your arrow keys
or mouse) to a different or new record. And in most instances, you will
be queried if you want to post any changes if you attempt to close a
form or move to a different tabbed page. But you must be aware that any
changes you make must be posted to the database, and that simply
changing a field will not by itself store this information.

The first thing to do when first setting up your company's scheduling
database is define when your business week begins. This is very
important to the scheduler because it is used to determine overtime.
Most company's weeks begin on Sunday at 00:00, and this is the default.

If you decide you're ready to start setting up your own database, make a
copy of BLANK.GDB (again, it's recommended you copy it to SCHDSOLV.GDB).
The first things to define are the basic tables (jobcodes, ratecodes,
and paygroups (locations, and events are just for storing accounting
information and are optional). Departments are also optional, but can be
useful to allowing selective grouping of employees. For example, you may
need to create a span in a schedule that selects workers from the OFFICE
department with jobcode TYPING, as opposed to any employee with the
TYPING jobcode. Note that you will not be able to assign a jobcode to an
employee without an associated ratecode - therefore, you will need to
define at least one jobcode.

The Pay Groups table deserves some explanation. Every employee must
belong to one (and only one) pay group. Pay groups are really just used
to store the simple overtime rules of the Pannet Scheduler. Max Hours is
the maximum number of hours an employee can work in one week before
they've worked overtime (note that it is the start time of a job, not
the end time, which determines which week a job's hours are accumulated
in). Break Time is the amount of time which should occur between shifts
(the end of one job and the start of another); if this amount of time
does not pass, overtime has occurred. Generally, Max Hours is 40 and
Break Time is 0.

Once you've entered the basic information needed, you can then enter
employee information. Simply click on the Add button in the employees
list window, and you can define a new employee. Note that you must
complete and post the Personal tabbed page for an employee before you
can enter their Availability or Jobs/Rates pages. Also note that an
employee must be made available sometime in the Sunday-Monday weekday
page, and they must be capable of performing some Jobcode, before they
will ever be potentially selected for a schedule.

Once you've entered in a few employees, try defining a schedule. The
most critical piece of information of a schedule is the start date. This
is because all spans (individual jobs) of a schedule are defined
relative to that start date: the day field for each span is the number
of days past that start date. This will likely be very confusing at
first (and maybe forever) - so experiment with it until you understand
it.

After you've defined the basic information of a schedule, and posted it,
you will then be able to input your spans. Spans can be thought of as
unique shift definitions - only one jobcode can be defined, but multiple
workers can be required. As stated, spans are defined relative to the
schedule's start date. Thus, if a schedule's start date is 01/01/1997
at 00:00, and a span's start day, time, and length are 0, 02:00, and 03:30
respectively, the span starts at 01/01/1997 02:00 and ends 3 1/2 hours
later. Likewise, if a schedule starts at 02/02/1998 at 05:00, and a span's
start day, time, and length are 3, 03:00 respectively, the span starts
at 02/05/1998 08:00 and ends three hours later.

Once you have defined a few spans, try clicking on the "Find Workers"
button on the Span Detail screen (or click on the "Find All Workers"
menu item under schedules when on the Schedule Detail screen - it finds
all the workers for all spans). While "finding" workers does not find a
solution to the schedule, it will give determine which workers could
possibly be chosen for a span. And if the scheduler engine
(SCHDSOLV.EXE) reports that not enough workers could be found, you can
be sure a solution will also not be found.

When you're reasonably sure enough workers are available for each span,
try finding a solution (click on the Find Solution menu while on the
Schedule Detail form). If the scheduler engine does find a solution (or
multiple - changing the "Max Solutions" value on the Schedule Detail
page will allow it to search for more than the default 1), you can then
review the solution by clicking on Show Solutions on the Schedule menu.

Finally, after you've found a solution to a schedule that is reasonably
close to what you want, commit that solution by clicking the Commit
button the Schedules Solution page. Committing a schedule locks that
solution in, changes the schedule itself to read-only, and creates
commitment records. A schedule can be de-committed by selecting it on
the Schedules page and clicking the Decommit button.

All commitments can be viewed on the Commitments form, either in textual
format or in a more visual format. No data can be changed here, except
you can select a specific record in the textual display and find
replacements for that worker at that time. You can reduce the number of
commitments you view by selecting specific schedules, workers, and/or
jobcodes and clicking the refresh button.

Schedule solutions and commitment records can be saved to file or
printed. In truth, at this time it is probably more useful to save your
commitments to a file than to print out the report because the
formatting of the reports is fairly lousy. But if you save your data to
a file, you can easily view data in a spreadsheet program. You will be
given the chance to define how the data is formatted.

That's all I can think to write right now. I hope you are able to get
some use of the system at this time, and please know a manual will be
available soon.

thank you,
Andrew Burke
Pannet Software
Portland, Oregon
aburke@pannet.com
503-232-7997