# How Zaphiri Uses Energy States
Energy-aware scheduling helps Zaphiri place demanding work where it has the best chance to fit while keeping routine work, breaks, and fixed commitments in view during [[Daily Planning]].
When you select **Optimize Schedule**, Zaphiri classifies flexible tasks, evaluates the constraints and evidence for the day, and shows a proposed schedule for your review.
> [!quote] Put the right work in the right part of the day
>
> Energy is one important scheduling signal. Zaphiri combines it with your priorities, commitments, routines, and deliberate planning choices.
## Automatic energy classification
Zaphiri classifies tasks as Peak, Flow, Steady, or Recharge. It can use the task's role and patterns from your own work, with deterministic local behaviour available when an AI provider is unavailable or returns an incomplete result.
Long or busy task lists are validated so a partial provider response does not silently leave the rest of the day unclassified.
> [!tip] You can override any label
>
> Change a task's energy state when the classification does not fit. Your correction persists when the schedule is rebuilt and takes priority over an automated suggestion.
## How schedule optimisation works
The scheduling engine considers several kinds of information without allowing any single signal to control the result:
- **Hard constraints**: Working hours, fixed calendar events, deadlines, unavailable time, dependencies, and minimum usable blocks
- **Current intent**: Priority, manual ordering, planned duration, Deep Work, and changes you make while planning
- **Task context**: Whether the work is preparation, execution, communication, administration, review, or closeout
- **Personal history**: Qualified patterns in recurring work, typical order, time of day, duration, and prior planning choices
- **Energy fit**: Your configured Peak, Flow, Steady, and Recharge periods
- **Schedule quality**: Usable gaps, transitions, capacity, breaks, and food-break settings
Fixed events stay protected. When a calendar provider fails, Zaphiri only uses eligible matching calendar context and marks the schedule as degraded. It does not silently remove a meeting to make the plan fit.
The result is a preview grouped into energy blocks, each labelled by time, energy state, and the tasks it contains.
![[energy-zones-schedule.png]]
Open a block to see its tasks. Expand **Optimization details** for the reasons behind consequential ordering decisions, the confidence of learned signals, and any work that could not fit into the legal capacity of the day.
## Learning from your work
Recurring tasks can build a pattern across previous instances. With enough consistent evidence, Zaphiri can use typical order, timing, and duration to make a schedule feel closer to your established routine.
- Recent consistent behaviour carries more weight than older behaviour
- One unusual day does not redefine the pattern
- Manual planning choices are stronger evidence than a schedule you only viewed
- Sparse, conflicting, or low-confidence history remains neutral
- History does not override fixed events or an explicit choice you made for today
Zaphiri keeps these signals scoped to your own permitted work. Team members do not receive access to your private Planning history or learned preferences.
## Example day
> [!example] How an explanation might read
>
> Its priority was the strongest scheduling signal. Fixed calendar events were kept in place.
## Review energy-profile suggestions
If a sufficiently strong pattern in your completed work differs from your configured focus window, Zaphiri can offer a suggestion for you to review. The card shows the observation period, completed-session count, and plain-language evidence.
Select **Use this window** to apply it, **Adjust** to edit it first, **Not now** to leave it for later, or **Dismiss this suggestion** when you do not want to use it. Zaphiri does not change your energy settings automatically, and it does not present an inferred pattern as a diagnosis.
---
**Next:** Review your own rhythm in [[Find Your Energy Pattern]], then put the schedule into action with [[Daily Planning]].