Node types
Process Map uses four node types — Start, Task, Queue, and End — that you connect together to model a workflow. This page explains what each type does and how to configure it.
Overview
Every process you model follows a path: work items enter through a Start node, move through Task nodes where processing happens (optionally waiting in Queue nodes when capacity is limited), and exit through an End node. Understanding when to use each type helps you accurately represent real-world workflows in the simulation.
Use a Queue node in front of any Task that has limited resources. This accurately models the backlog that builds when demand exceeds capacity, and lets you measure wait times in simulation results.
Core concepts
| Concept | Description |
|---|---|
| Work item | A single unit that flows through the process (a customer, a ticket, a job) |
| Arrival | When a new work item enters the process at the Start node |
| Service time | How long a Task takes to process one work item |
| Resource | A shared capacity constraint (for example, a staff member) assigned to a Task |
| Queue | A buffer where work items wait when downstream capacity is unavailable |
A well-formed process must have exactly one Start node and at least one End node.
Start
The Start node is the entry point for work items. You configure it with distributions that control how often items arrive and how much work each arrival brings.
Every scenario must have exactly one Start node. The simulation will not run if you have zero or more than one.
Fields:
| Field | Description |
|---|---|
| Label | Name displayed on the canvas |
| Arrival frequency distribution | How often work items arrive (for example, exponential(rate=2) means on average 2 arrivals per time unit) |
| Arrival severity distribution | Optional quantity per arrival |
Task
A Task node represents a processing step. When a work item arrives, it spends time being processed — sampled from the service time distribution you configure — before moving to the next node.
Fields:
| Field | Description |
|---|---|
| Label | Name displayed on the canvas |
| Distribution | Service time distribution |
| Resources | Optional resource assignments (limits concurrency) |
Queue
A Queue node is a waiting room. Work items accumulate here until capacity is available downstream. Use it to model buffers and backlogs in your process.
Fields:
| Field | Description |
|---|---|
| Label | Name displayed on the canvas |
| Capacity | Maximum items that can wait simultaneously (null = unlimited) |
Queues do not consume service time — items pass through instantly once capacity becomes available.
End
The End node is the exit point for work items. Every scenario must have at least one End node.
If your process has no End node the simulation cannot run. Add at least one End node to complete the flow.
Node indicators
Each node on the canvas shows visual indicators so you can see its configuration at a glance.
| Indicator | Meaning |
|---|---|
| Distribution badge | Shows the configured distribution type |
| Resource dots | Colored dots for each assigned resource |
| Sim icon | Opens the simulation config panel for this node |
| Info icon | Opens the node info panel |
How to configure a Task node
- Click a Task node on the canvas to select it. The node panel opens on the right.
- Enter a descriptive Label (for example, "Review application").
- Choose a Distribution for the service time. This controls how long each work item spends at this step.
- Optionally assign one or more Resources to limit how many items can be processed simultaneously.
- Click outside the panel or press Escape to close it. The node updates to reflect your configuration.
How to add a Queue before a constrained Task
Use this pattern whenever a Task has limited resources and you expect demand to exceed capacity.
- Click the Queue node type in the toolbar and place it on the canvas before the Task you want to buffer.
- Click the Queue node and enter a Label (for example, "Waiting for review").
- Set Capacity to
nullfor an unlimited buffer, or enter a number to cap the queue size. - Draw a connection from the upstream node to the Queue, then from the Queue to the Task.
- Run the simulation. The avg queue length and avg wait time metrics for the Queue node show the backlog effect.