Considerations for Using the Dispatcher Agent in Field Service
Keep these considerations in mind when using the dispatcher agent to fill schedule gaps, get a daily summary of scheduling issues, or filter your appointment list.
Required Editions
| Einstein generative AI is available in Lightning Experience. Setup for Einstein generative AI is available in Lightning Experience. |
| Available in all enhanced Messaging channels. View required editions. |
| The Field Service core features and managed package are available in Enterprise, Performance, Unlimited, and Developer Editions. |
| For purchasing options, contact your Salesforce account executive. |
Considerations for Filling Gaps
- If a resource’s shift spans over 2 days, the AI agent can show gaps and schedule appointments that begin on the requested day and extend into the next day.
- The AI agent uses the territory’s time zone to look for gaps and service appointments. However, when communicating with the dispatcher, the AI agent uses the time zone defined in the Classic Dispatch Console’s settings, either user or territory.
- The AI agent shows gaps for 1 day and doesn’t support time ranges.
- If the user doesn’t specify a specific date, the AI agent searches for gaps and recommends appointments for today.
- The AI agent shows appointments for the first gap that has service appointments available to fill it. If there’s an earlier gap in the day but the AI agent can’t find service appointments to fill the gap, it doesn’t offer this gap, and instead displays the first gap that has appointments to fill it.
- Service appointments that are part of a complex work chain, span over multiple days, or are pinned, aren't considered as eligible to fill the gap.
- When you’re searching for service appointments or scheduling them, no reshuffling or sliding is performed.
- Service appointments are filtered out and analyzed based on the scheduling policy selected in the Gantt. If the Gantt is closed, the Gantt’s default policy is used.
- The AI agent prioritizes finding unscheduled appointments. Only when no unscheduled
appointments are available does it proceed to locate the most suitable scheduled
appointments that aren’t scheduled for the requested date.
Scheduled appointments that are customer-facing and have arrival windows aren't considered as eligible to fill gaps. Unscheduled appointments with arrival windows are considered.
- To define what is considered to be a gap, the AI agent uses the predefined estimated job duration for work types. In addition, resource efficiency is considered. If work types aren’t configured, the AI agent examines the working territories of the resource to identify the shortest job duration from the previous week while considering the resource’s efficiency.
- The time horizon of appointments offered to customers is the earliest date between:
- the appointment's due date, and
- the value defined in Maximum days to get candidates or to book an appointment. To find this setting, go to the Field Service Settings app and open the General Logic tab. The maximum value is 31. The default value is 10.
For example, if the due date is in 60 days and the maximum days to get candidates or to book an appointment is set to 15, the customer is offered appointments within 15 days. Or, if the due date is in 20 days and the maximum days to get candidates or to book an appointment is set to 30, the customer is offered appointments within 20 days.
- The ability to fill gaps is possible only when Enhanced Scheduling and Optimization is enabled.
Limitations for Filling Gaps
- When a resource has two shifts on two consecutive days, and the travel time to an appointment on day two starts on day one, this travel time is assigned to day two and therefore the AI agent isn’t aware of it when searching for gaps on day one.
- Contractors aren’t supported, as Field Service doesn’t manage their schedules or routes.
Considerations for Getting Daily Summaries
- If you ask the AI agent for a schedule summary, you get a summary for all territories visible in the Gantt. If you ask for a schedule summary for all territories, you get a summary for all territories you have access to.
- If more than one Gantt is open, and you ask the AI agent for a schedule summary, you get a summary based on the territories in the last Gantt you opened.
- When you ask for a schedule summary, the summary includes appointments for territories visible in the current Gantt only. For example, John Smith has appointments in Western New York and Central New York, but you’re Gantt shows Western New York only. When you ask for a schedule summary, you see a summary for appointments in Western New York only.
- To get a schedule summary for specific service territories, you must first search for appointments from those service territories, and only then ask to summarize the results. For example, to get a summary of scheduling issues for all territories containing the word New, you first need to search for the appointments. For example, type Find me appointments for territories containing New, and after you see the appointments you can ask for a summary by typing Create a schedule summary for these appointments.
Limitations for Getting Daily Summaries
- The Appointments with Rule Violations and Overlapping Appointments categories are available for Enhanced Scheduling and Optimization only and are based on up to 4,000 records.
- If the Gantt is closed, the summary is created but the filters aren’t.
- You can’t remove or change the categories named Appointments with Rule Violations and Overlapping Appointments.
- Agentforce can’t generate summaries if more than 20 territories are selected in the Classic Dispatch Console.
- If Enhanced Scheduling and Optimization is enabled for only some of the selected territories, the Appointments with Rule Violations and Overlapping Appointments categories aren’t shown.
Considerations for Filtering Appointment Lists
- This feature works only when the Gantt is open, even if the user receives a message that the filter was created when the Gantt is closed.
Limitations for Filtering Appointment Lists
- This feature is dependent on the QueryRecords action to search for service appointments, and is subject to the QueryRecords limit of returning a maximum of 25 records.
Did this article solve your issue?
Let us know so we can improve!
