Understand the result table parameters

The parameters in the Result table in the Schedules module are important to understand when reviewing the result of the scheduling and optimization.

Which parameters that are shown in the Result table vary with the type of skill selected and depending on if you are looking at the table on intraday level or on, for example, the day level. There are also parameters that are only shown when there are activities that are shorter than an interval or that start or end within an interval.

The values in the table can be shown with or without shrinkage. Select the Shrinkage button in the ribbon to have the results factor in the shrinkage estimates that are entered in the Forecast.

EXAMPLE   The shrinkage for the workload is set to 5%. Select the Shrinkage button to take the 5% cushion into consideration and show the matching results. Click it again to remove the 5% cushion and show the staffing with just the raw staffing calculation.

Forecasted hours or agents

Forecasted hours is the number of agent hours needed to cover the forecasted workload. On an intraday level, Forecasted agents is used instead. This is the number of agents needed to handle the tasks forecasted for each interval. These values are loaded from the detailed forecast for the skills.

For channel types like email and back office, the Forecasted hours or agents value is affected by what is scheduled and then the values do not match what is in the original forecast for that day or interval. This is because those tasks do not have to be handled in the same interval that they arrive. They can be handled from the interval that they arrive and in any of the following intervals, within the Handled within time. Therefore, the work can be handled earlier or later than what was planned in the forecast, depending on when more agents are available based on the schedule.

EXAMPLE   

A back office skill has the Handled within target defined as 4h. The interval length is 1h.

Forecasted volume, before scheduling.

Time 09:00 10:00 11:00 12:00

Forecasted agents

1:00

1:00

1:00

1:00

Scheduled agents

0:00

0:00

0:00

0:00

Absolute difference

-1:00

-1:00

-1:00

-1:00

Relative difference

-100%

-100%

-100%

-100%

Forecasted volume after scheduling, with 1h understaffing.

Time 09:00 10:00 11:00 12:00

Forecasted agents

0:30

1:00

1:15

1:15

Scheduled agents

0:15

0:45

1:00

1:00

Absolute difference

-0:15

-0:15

-0:15

-0:15

Relative difference

-50%

-25%

-20%

-20%

The Forecasted agents value is adjusted to keep the Absolute difference even across the Handled within time intervals.

Scheduled hours, agents, or heads

Scheduled hours is the number of agent hours scheduled on the selected skill. On the intraday level, Scheduled agents is used instead. This is the number of agents (FTEs) scheduled to handle the tasks for each interval.

Scheduled heads is the number of actual persons scheduled for a skill. This is only shown on intraday level.

EXAMPLE   9 multi-skilled agents are scheduled to work on a skill during an interval. As they are also taking calls on other skills, the work they are scheduled to do on this skill is in total as if 4 agents were working only on this skill. In this case, Scheduled agents shows 4 and Scheduled heads shows 9.

If there are limits on minimum or maximum agents for a skill and those limits are broken, that is highlighted on the Scheduled hours row. If the minimum agents limit is not fulfilled, the value in the Scheduled hours row is highlighted in red. If the maximum agents limit is exceeded, the value in the Scheduled hours row is highlighted in blue. On an intraday level, it is the Scheduled heads value that is highlighted if a maximum or minimum agents limit is broken.

Limits on maximum and minimum agents are defined in the skill targets.

Absolute, relative, and adjusted difference

The Absolute difference presents the difference between what is forecasted and what is scheduled in number of hours or, on intraday level, in number of agents.

The Relative difference presents the difference between what is forecasted and what is scheduled as a percentage.

When the Relative difference exceeds the understaffing and overstaffing thresholds that are set for the skill they are highlighted. Red means critical understaffing, yellow means understaffing and blue means overstaffing. This helps you quickly identify days or intervals where too few or too many agents are scheduled.

EXAMPLE   

Skill properties are set up to have Overstaffing start at 20%. The skill properties are set up to have Understaffing start at -10% and Critical understaffing start at -25%.

  • 20% to 100% are highlighted in blue.
  • 20% to –10% are not highlighted.
  • -10% to –25% are highlighted in yellow.
  • -25% to –100% are highlighted in red.

If the Relative difference for a day is highlighted in red in the Day view (right-click in the Results view to change your view), even though the skill is overstaffed on day level, that means that there is at least one critically understaffed interval within the day. If the forecast for an interval is zero, the relative difference cannot be calculated. The Relative difference is in these cases displayed as NaN.

During optimization of the schedule, the optimization uses the Relative difference to identify the days or intervals with the highest and lowest relative staffing with the purpose of achieving a more even staffing level for the whole period. In the rare cases that Tweaked values are used when optimizing, the optimization uses the Adjusted difference instead.

RMS and standard deviation

RMS (Root mean square) and Standard deviation are useful values to look at to understand the distribution of the scheduled agents compared to the forecast.

The RMS value tells you how close the staffing follows the forecast within that period. Standard deviation tells you how evenly the scheduled hours are distributed according to the forecast levels. Primarily, strive to achieve even levels for all days and intervals of the schedule period. Secondly, strive to achieve as low values as possible.

EXAMPLE   

There are 4 intervals in a day. The Relative difference values for these intervals affect the RMS and Standard deviation values on day level.

  • In situation 1, the schedule perfectly matches the forecast, and both RMS and Standard deviation are 0.
  • In situation 2, the skill is understaffed, but evenly understaffed for all intervals. RMS is 10, which means that the skill is understaffed or overstaffed. Standard deviation is 0, as the scheduled hours are evenly distributed between the intervals.
  • In situation 3, the staffing of the skill is not close to the forecast, as shown by the RMS. The staffing is very unevenly distributed, indicated by the Standard deviation.

 

Situation

Relative Difference

Interval 1

Relative Difference

Interval 2

Relative Difference

Interval 3

Relative Difference

Interval 4

 

RMS

Standard

Deviation

1

0

0

0

0

0

0

2

-10

-10

-10

-10

high

0

3

-10

10

-10

10

high

high

Predicted service level

Predicted service level (PSL) indicates the service level that can be reached with the current schedules. PSL is calculated based on the forecast, considering the efficiency and the occupancy factors. The shrinkage factor is taken into consideration when calculating PSL if the Shrinkage button is selected. It is important to note that the Predicted service level is, in fact, a prediction and not a guarantee of performance.

PSL is calculated differently for different skill types.

  • For inbound telephony, chat, and retail skills, PSL is calculated with Erlang A. It shows the percentage of calls or chats that is predicted to be handled within the number of seconds defined in the service level target.
  • For the other skill types, PSL is calculated as Scheduled agents incoming divided by Forecasted agents incoming.

The PSL is the same as the defined service level target if the number of scheduled agents equals the number of forecasted agents. The highest possible PSL percentage is 100%.

EXAMPLE   For one interval the target service level is 80% answered within 20s. Considering volumes and handling time, this leads to a forecasted staff of 10 agents. When 10 agents are scheduled, the PSL is 80%. This means that 80% of the calls are predicted to be answered within the 20s defined in the target service level.

Predicted average speed of answer

Predicted average speed of answer (PASA) indicates the average speed of answer that can be reached with the current staffing levels. PASA is calculated based on the forecast, considering the efficiency and the occupancy factors. The shrinkage factor is taken into consideration when calculating PASA if the Shrinkage button is selected.

PASA is calculated differently for different channel types.

  • For inbound telephony and chat channel types, PASA is calculated with Erlang A. It shows the predicted average time a customer must wait in queue.
  • For the other channel types, PASA is calculated with this formula:
    PASA = Forecasted agents incoming × Handled within ÷ (2 × Scheduled agents incoming)

NOTE   In a skill summary where different channel types are included, PASA is calculated for just the telephony and chat skills.

Predicted abandon rate

Predicted abandon rate (AR) indicates the abandon rate that can be reached with the current staffing levels. Predicted AR is calculated based on the forecast and scheduled agents, considering the efficiency and the occupancy factors. The shrinkage factor is taken into consideration when calculating predicted AR if the Shrinkage button is selected.

Predicted AR is only calculated for inbound telephony and chat channel types. It is calculated with Erlang A and shows the predicted abandon rate for a customer.

NOTE   In a skill summary where different channel types are included, Predicted AR will only be shown if all underlying skills are telephony or chat skills

Forecasted and scheduled hours incoming

The Forecasted hours incoming and Scheduled hours incoming are used for email and back office skills to describe the total forecasted and scheduled hours for the tasks or emails that arrive on a day. On intraday level, Forecasted agents incoming and Scheduled agents incoming are used instead.

When you plan for back office or email skills, the tasks or emails arrive at one point in time but can normally wait longer than a call to be handled. The agents can be scheduled to handle the work that arrives in one interval over the following intervals, according to the defined Handled within target on the skill.

EXAMPLE   

A back office skill has all incoming volume within the first interval. The Handled within target is defined as 4 h. The interval length is 1 h.

Time 09:00 10:00 11:00 12:00

Forecasted agents

0:30

1:00

1:15

1:15

Scheduled agents

0:15

0:45

1:00

1:00

Forecasted agents incoming

4:00

0:00

0:00

0:00

Scheduled agents incoming

3:00

0:00

0:00

0:00

Absolute and relative difference incoming

The Absolute difference incoming and Relative difference incoming present the difference between what the Forecasted hours incoming and the Scheduled hours incoming in actual numbers and as a percentage.

Intra interval balance

The Intra interval balance indicate how well the staffing is balanced within each interval. The Lowest intra interval balance value presented on day level is the lowest Intra interval balance value for an interval on that day.

The Intra interval balance is shown for telephony and chat skills, but only show values when there are activities that are shorter than an interval or that start or end within an interval. Intra interval resources are calculated for each 5-minute period within the interval.

The balance for an interval is calculated by dividing the lowest number of resources for a 5-minute period with the highest number of resources for a 5-minute period. If the Intra interval balance is 100%, it is perfectly balanced. If the value is less than 80%, this is considered to be an issue and is highlighted in red.

EXAMPLE   

In one interval, 100 agents are scheduled to work on a phone skill. They all have a 5-minute break within the interval.

  • 50 of the agents have their break the first 5-minute period.
  • 30 of them have their break the second 5-minute period.
  • 20 of them have their break the last 5-minute period of the interval.

The intra interval balance is 62,5%, since the minimum resources is 50 when the other 50 agents are on break and the maximum resources is 80, when only 20 agents are on break.

To solve issues with the Intra interval balance, use the Intra interval balance optimization step.

Related topics