> ## Documentation Index
> Fetch the complete documentation index at: https://documentation.kodelabs.com/llms.txt
> Use this file to discover all available pages before exploring further.
# Create an M&V model
> Configure and train measurement and verification models for energy savings analysis
Configure and train Measurement and Verification (M\&V) models at the building level to verify energy savings. In the [building view](/products/energ/get-started/navigation#energy-modeling-submenu), expand **Energy Modeling** in the left sidebar and open **M\&V**. A trained model estimates the adjusted baseline and feeds savings calculations in KODE OS and the [Building Energy dashboard](/products/energ/analytics/building-dashboards-energy#mv). For the methods behind M\&V, see [Energy modeling and M\&V](/products/energ/concepts/energy-modeling).
You typically need about 12 months of historical data in the baseline window so weather and seasonal patterns train reliably.
## Open M\&V
From [Portfolio Home](/products/energ/analytics/benchmarking), open a building in the building view.
In the left sidebar, expand **Energy Modeling**, then click **M\&V**.
## Manage configurations
The **M\&V** page lists every configuration for the building. Click **+ Create Configuration** to start the five-step wizard. Use the **Actions** menu on a row to edit or delete an existing configuration.
| Column | What it shows |
| ---------------------- | -------------------------------------------------------------------------------------------------------------- |
| **Name** | Configuration label. The **Default** tag marks the configuration EnerG uses when you open M\&V from dashboards |
| **Meter Category** | **Utility** for bill-based meters or **Interval** for subhourly reads |
| **Aggregation** | Rollup level, such as **Monthly** |
| **Meters** | Count of meters included in the configuration |
| **Baseline Period** | Training window start and end dates |
| **Configuration Type** | **Auto Configure** or **Advanced** |
| **Status** | Training state, such as **Succeeded** |
| **Actions** | Row menu for edit, delete, or related tasks |
## Create an M\&V configuration
Click **+ Create Configuration** to open the wizard. Progress appears in the left sidebar across five steps: **Basic Setup**, **Data Selection**, **Baseline Setup**, **Model Configuration**, and **Summary**.
### Step 1: Basic setup
Define the configuration name, optional description, and measure dates that mark when savings counting begins.
| Field | Purpose |
| ----------------------------- | ----------------------------------------------------------------------------------------------------------------------- |
| **Name** | Label shown in lists and dashboard filters |
| **Default** | When checked, EnerG selects this configuration on M\&V dashboards |
| **Description** | Optional notes for auditors or internal teams |
| **Measure Start Period** | **Get from existing Measure** pulls dates from Scenario Planner, or **Choose Measure Dates** lets you set them manually |
| **Implementation Start Date** | When the retrofit or measure work began |
| **Operation Start Date** | When the measure reached normal operation. Charts show this as the **Operation Date** marker |
Click **Next** to continue.
### Step 2: Data selection
Choose meter category, aggregation, and the meters that feed the model.
| Field | Purpose |
| ------------------ | ----------------------------------------------------------------------------------------- |
| **Meter Category** | **Utility** for monthly billing meters or **Interval** for high-frequency reads |
| **Aggregation** | Granularity for interval data, such as **Monthly**, **Daily**, or **Hourly** |
| **Meter list** | Search and filter by **Meter Scope** and **Service Type**, then select one or more meters |
EnerG trains a separate model per service type when you include multiple commodities, such as **Electric** and **District Steam**.
### Step 3: Baseline setup
Select the historical period used to train the baseline and set degree-day base temperatures.
| Field | Purpose |
| ----------------------------- | ------------------------------------------------------------------------------------------------------------------- |
| **Historical Usage Trends** | Preview monthly use by service type before you lock the baseline window |
| **Baseline Period** | **Start Date** and **End Date** for training, such as **01/01/2022 – 12/31/2022** |
| **Building Base Temperature** | **Get from Building Information** or **Choose Base Temperature** for custom **Cooling** and **Heating** base values |
Confirm the chart shows a full seasonal cycle before you click **Next**.
### Step 4: Model configuration
Pick how EnerG trains the regression models and whether to project the baseline forward.
#### Auto-configure
**Auto-Configure** trains and selects the optimal model for each service type without manual variable selection.
| Field | Purpose |
| ----------------------- | -------------------------------------------------------------------- |
| **Configuration Type** | **Auto-Configure** or **Advanced** |
| **Disable Projection** | When on, EnerG skips forward baseline estimates |
| **Baseline Projection** | Horizon from one to 12 months for future adjusted-baseline estimates |
#### Advanced
**Advanced** exposes drivers, model type, and accuracy thresholds. Use the scatter plot to confirm relationships before you submit.
| Field | Purpose |
| ------------------------------------- | ------------------------------------------------------------------------------------------- |
| **Select the Input Variable to Plot** | **Temperature**, **Humidity**, **Pressure**, or **Wind Speed** for the preview chart |
| **Scatter Plot** / **Historical** | Toggle between correlation view and time-series view |
| **Variables** | Drivers such as degree days, holidays, or interval-of-week for interval meters |
| **Model Type** | Regression family, filtered by enabled variables |
| **Threshold Percentage** | **Model Metrics** uses MAPE as the threshold, or **Set the Threshold** for a custom percent |
### Step 5: Summary
Review selections and trained model status before you finish.
| Section | What it shows |
| ------------------------------- | ---------------------------------------------------------------------------------------------------------------------------- |
| **Baseline Period** | Confirmed training window |
| **Meters Included vs Selected** | How many meters trained versus how many you selected |
| **Configuration Type** | **Auto Configure** or **Advanced** |
| **Trained Models** | Per-service-type result, such as **Electric Model** and **District Steam Model**, each with **Successful** or failure detail |
Click **Finish** to save the configuration. Results sync to dashboards after the scheduled job runs. See [M\&V analysis](/products/energ/energy-modeling/mv-analysis) to read savings totals, charts, and accuracy metrics.
Interval models usually take longer to process. Auto-configure may test multiple candidates before it returns a result.
## When models fail
Models can fail when correlation is too weak or the fit falls below your accuracy threshold. Interval training may also reject configurations that do not meet quality rules. See [Model accuracy thresholds](/products/energ/reference/calculations-and-parameters#model-accuracy) for the metrics EnerG checks.
If a run fails, widen the baseline, adjust variables, or verify [meter completeness](/products/energ/buildings/meters) before you retry.
## Next steps
Interpret savings, charts, and accuracy metrics after the hourly job finishes.
Review M\&V savings and consumption charts without leaving the Energy dashboard.
See how IPMVP options, meter types, and forecasting fit together in EnerG.
Configure horizons and read projected consumption alongside actual use.
| Column | What it shows |
| ---------------------- | -------------------------------------------------------------------------------------------------------------- |
| **Name** | Configuration label. The **Default** tag marks the configuration EnerG uses when you open M\&V from dashboards |
| **Meter Category** | **Utility** for bill-based meters or **Interval** for subhourly reads |
| **Aggregation** | Rollup level, such as **Monthly** |
| **Meters** | Count of meters included in the configuration |
| **Baseline Period** | Training window start and end dates |
| **Configuration Type** | **Auto Configure** or **Advanced** |
| **Status** | Training state, such as **Succeeded** |
| **Actions** | Row menu for edit, delete, or related tasks |
## Create an M\&V configuration
Click **+ Create Configuration** to open the wizard. Progress appears in the left sidebar across five steps: **Basic Setup**, **Data Selection**, **Baseline Setup**, **Model Configuration**, and **Summary**.
### Step 1: Basic setup
Define the configuration name, optional description, and measure dates that mark when savings counting begins.
| Field | Purpose |
| ----------------------------- | ----------------------------------------------------------------------------------------------------------------------- |
| **Name** | Label shown in lists and dashboard filters |
| **Default** | When checked, EnerG selects this configuration on M\&V dashboards |
| **Description** | Optional notes for auditors or internal teams |
| **Measure Start Period** | **Get from existing Measure** pulls dates from Scenario Planner, or **Choose Measure Dates** lets you set them manually |
| **Implementation Start Date** | When the retrofit or measure work began |
| **Operation Start Date** | When the measure reached normal operation. Charts show this as the **Operation Date** marker |
Click **Next** to continue.
### Step 2: Data selection
Choose meter category, aggregation, and the meters that feed the model.
| Field | Purpose |
| ------------------ | ----------------------------------------------------------------------------------------- |
| **Meter Category** | **Utility** for monthly billing meters or **Interval** for high-frequency reads |
| **Aggregation** | Granularity for interval data, such as **Monthly**, **Daily**, or **Hourly** |
| **Meter list** | Search and filter by **Meter Scope** and **Service Type**, then select one or more meters |
EnerG trains a separate model per service type when you include multiple commodities, such as **Electric** and **District Steam**.
### Step 3: Baseline setup
Select the historical period used to train the baseline and set degree-day base temperatures.
| Field | Purpose |
| ----------------------------- | ------------------------------------------------------------------------------------------------------------------- |
| **Historical Usage Trends** | Preview monthly use by service type before you lock the baseline window |
| **Baseline Period** | **Start Date** and **End Date** for training, such as **01/01/2022 – 12/31/2022** |
| **Building Base Temperature** | **Get from Building Information** or **Choose Base Temperature** for custom **Cooling** and **Heating** base values |
Confirm the chart shows a full seasonal cycle before you click **Next**.
### Step 4: Model configuration
Pick how EnerG trains the regression models and whether to project the baseline forward.
#### Auto-configure
**Auto-Configure** trains and selects the optimal model for each service type without manual variable selection.
| Field | Purpose |
| ----------------------- | -------------------------------------------------------------------- |
| **Configuration Type** | **Auto-Configure** or **Advanced** |
| **Disable Projection** | When on, EnerG skips forward baseline estimates |
| **Baseline Projection** | Horizon from one to 12 months for future adjusted-baseline estimates |
#### Advanced
**Advanced** exposes drivers, model type, and accuracy thresholds. Use the scatter plot to confirm relationships before you submit.
| Field | Purpose |
| ------------------------------------- | ------------------------------------------------------------------------------------------- |
| **Select the Input Variable to Plot** | **Temperature**, **Humidity**, **Pressure**, or **Wind Speed** for the preview chart |
| **Scatter Plot** / **Historical** | Toggle between correlation view and time-series view |
| **Variables** | Drivers such as degree days, holidays, or interval-of-week for interval meters |
| **Model Type** | Regression family, filtered by enabled variables |
| **Threshold Percentage** | **Model Metrics** uses MAPE as the threshold, or **Set the Threshold** for a custom percent |
### Step 5: Summary
Review selections and trained model status before you finish.
| Section | What it shows |
| ------------------------------- | ---------------------------------------------------------------------------------------------------------------------------- |
| **Baseline Period** | Confirmed training window |
| **Meters Included vs Selected** | How many meters trained versus how many you selected |
| **Configuration Type** | **Auto Configure** or **Advanced** |
| **Trained Models** | Per-service-type result, such as **Electric Model** and **District Steam Model**, each with **Successful** or failure detail |
Click **Finish** to save the configuration. Results sync to dashboards after the scheduled job runs. See [M\&V analysis](/products/energ/energy-modeling/mv-analysis) to read savings totals, charts, and accuracy metrics.