Use AI Scheduling Suggestions
When an advisor or controller needs to assign a technician to a job, BayWise can offer a ranked recommendation based on skill match, current workload, and historical performance on similar job types. The recommendation comes back in seconds with a short plain-language reason, so the advisor can make an informed decision quickly — especially during a busy morning when the board is filling up.
This guide explains how to use the AI Suggest feature, what drives the recommendation, and when to trust it or override it.
Prerequisites: The Scheduling Suggestions AI function must be configured. See Configure AI Providers.
What the AI considers
When you request a suggestion for a job, BayWise assembles a summary of the job and the available technicians, then sends that summary to your configured AI provider. The AI evaluates:
- Skill match — does the technician have the skills required for this job’s service type?
- WIP limit — how many jobs is the technician currently handling, and are they within their maximum work-in-progress limit?
- Workload balance — how does this technician’s current load compare to the rest of the team?
- Historical efficiency — on completed jobs of this service type, what was this technician’s average efficiency rating?
The AI returns a ranked recommendation — typically the top choice with a plain-language explanation, and sometimes a second-best option if the top choice has a caveat (such as being close to their WIP limit).
Example output:
Priya Nair — Skill match: Master Technician, Diagnostics. Current WIP: 1 of 3. Efficiency on similar jobs: 94%. Recommended.
Alternative: Carlos Mendoza — Skill match: Master Technician. WIP: 2 of 3. Efficiency on similar jobs: 88%. Consider if Priya is unavailable.
The advisor reads this, confirms it makes sense, and either accepts or overrides the suggestion.
AI suggestions are recommendations, not commands. The advisor always makes the final assignment. Use the suggestion as a starting point — if your instinct differs from the AI (for example, you know a technician has a particular strength with this vehicle make, or they have agreed to take on an extra job today), trust your knowledge of your team.
How to use AI Suggest
Open a job that needs a technician assigned
From the Dashboard or the Calendar, find a job that has no technician yet — or a job where you want to reconsider the current assignment.
Click the job card to open the job detail panel.
Open the assignment panel
In the job detail panel, locate the Technician field. If no technician is assigned, you will see an Assign button. If a technician is already assigned, you will see their name with an Edit option.
Click Assign (or Edit) to open the assignment panel.
The assignment panel shows:
- A list of all available technicians for the current date and location
- Each technician’s current WIP count
- Each technician’s relevant skills (highlighted if they match the job’s service requirements)
Click AI Suggest
In the assignment panel, click the AI Suggest button (star icon, or labelled “Suggest”).
BayWise sends the job context to your configured AI provider. This takes approximately 2–5 seconds depending on the provider and model.
Read the suggestion
The suggestion appears directly below the AI Suggest button. It shows:
- The recommended technician’s name
- The reasons for the recommendation (skill match, WIP status, efficiency score)
- If applicable, an alternative technician with a brief note
Review the reasoning. Confirm it makes sense given what you know about the day’s situation — appointments, technician availability, vehicle-specific nuances.
Accept or override
To accept the suggestion: Click Assign [Technician Name] to confirm the assignment. The technician is assigned to the job. The job appears on their calendar slot.
To choose a different technician: Click any other technician in the list. The AI suggestion is not applied. Your manual choice is saved.
To dismiss and decide later: Close the assignment panel. The job remains unassigned. You can return to it at any time.
When AI scheduling works best
AI suggestions become progressively more useful as your BayWise data matures. The model has more to work with when:
- Your workshop has at least 2–3 weeks of completed job data. The efficiency scores used in the recommendation are based on completed jobs; with no history, the AI can only use skill match and WIP.
- Your service catalog has skill requirements set on each service. Without this, the AI cannot determine whether a technician’s skills are relevant to the job.
- Your technicians have accurate, up-to-date skill profiles. A technician with outdated skills listed will receive less relevant suggestions.
- Your technicians have WIP limits configured. Without WIP limits, the AI cannot factor in workload balance.
A well-maintained BayWise setup — accurate skills, complete catalog, a few weeks of history — gives the AI enough signal to produce recommendations your team will find genuinely useful.
When AI may not help
The suggestion will be weak or less reliable in these situations:
- First week of use. No completed jobs means no efficiency data. The AI can still reason about skill match and WIP, but the recommendation is thin.
- Skill requirements not set in the catalog. If the job’s service type has no required skills configured, skill match cannot be evaluated. All technicians appear equally qualified to the AI.
- All technicians at WIP limit. If every available technician is at their maximum work-in-progress, the AI will note this and may recommend the one closest to their limit — but the advisor should treat this as a staffing capacity issue, not a scheduling problem to solve with AI.
- Small team (2–3 technicians). With a very small team, manual assignment is fast and the advisor already knows each technician’s situation. AI suggestions add less value here than at workshops with 8–15 technicians.
Interpreting the reasoning
The AI gives a short reason for each recommendation. Here is how to read the key terms:
| Term in suggestion | What it means |
|---|---|
| Skill match: Master Technician, Diagnostics | The technician has both skills that this service type requires |
| Current WIP: 2 of 4 | The technician has 2 active jobs against a configured limit of 4 |
| Efficiency on similar jobs: 91% | On completed jobs of this service type, this technician completed them at 91% of the estimated duration on average |
| WIP at limit | This technician is at or above their configured WIP ceiling — assigning another job is possible but will exceed their limit |
| Partial skill match | The technician has some but not all of the required skills — the AI is suggesting them because no better match is available |
If the AI says “partial skill match” or “WIP at limit”, treat the suggestion as a fallback rather than a strong recommendation and review the alternatives carefully.
Multi-step jobs and AI Suggest
For multi-step jobs (collision repair, full mechanical rebuild), each step can have its own technician. The AI Suggest button is available per step in the step assignment panel. When suggesting for a step, the AI considers:
- The skill required for that specific step (e.g., “Paint Application” requires a Paint skill)
- Which technicians are already assigned to other steps of the same job (to flag potential conflicts)
- Each technician’s WIP at the time the step is scheduled to start
You do not need to assign all steps at once. It is common practice to assign Step 1 immediately and then assign later steps closer to when they are due to start.
Common questions
The AI Suggest button is greyed out. Why? The Scheduling Suggestions AI function is not configured, or the configuration has an error. Go to Settings → AI → Scheduling Suggestions and check the card status. See Configure AI Providers.
The suggestion recommended a technician who is off today. Why? The AI works from technician profiles and job data. If a technician’s availability for today has not been updated in BayWise (for example, an unplanned absence), the AI does not know they are out. Always cross-check the suggestion against your actual team availability on the day.
The AI keeps recommending the same technician for every job. Is that normal? Early in your BayWise use, one technician may appear to be the strongest match based on skills alone. As more job history accumulates, efficiency data will differentiate between technicians more clearly. If it continues, check that all technicians have their skills and WIP limits configured correctly — a technician with no skills entered will never appear as a strong match, and the AI will keep routing to those who do.
Can I see the full reasoning behind a suggestion? The brief explanation in the assignment panel covers the main factors. There is no expanded view of the AI’s full reasoning chain in the current version.
Does the AI account for technician preferences (e.g., a tech who prefers not to do diesel)? Not directly. Technician preferences are not a configured input in the current model. The AI reasons about skills, WIP, and efficiency. If a technician should not be assigned a certain job type, remove that service’s required skill from their profile, or use a service catalog flag to restrict compatible technicians.
If I always override the AI, does BayWise learn from that? Not in the current version. BayWise does not retrain the AI model based on your override choices. The AI suggestions improve naturally as your completed job data grows, but it does not adapt to your personal override patterns.