# Infinite-capacity scheduling

Use unconstrained dates to expose required production load, while keeping calendars, resource qualifications and executable capacity distinct.

Reviewed: 2026-11-01

Tags: production, planning, scheduling, capacity

In the worked example, one machine has **360 min/day** and three jobs each need **180 exclusive machine minutes** with no setup. An infinite schedule books all three at **08:00–11:00** on day 1, placing **540 minutes** of work against **360 minutes** of capacity. **150% loading** and a **180-minute shortfall** describe demand on the calendar. A feasible schedule needs different dates, capacity or another approved change.

ETH Zürich distinguishes infinite loading, which compares required load with capacity after allocation, from finite loading, which considers capacity before placing work. [Paul Schönsleben, ETH Zürich, section 5.3.4](https://opess.ethz.ch/course/section-5-3/5-3-4-overview-of-scheduling-and-capacity-management-techniques/)

Resolve overloaded dates before releasing work or committing delivery.

## Define which constraint is relaxed

Infinite capacity does not necessarily mean zero processing time, ignored routing or a 24-hour calendar. Infor's **CSI 2026.10** Scheduler documentation explicitly allows an infinite resource to perform unlimited simultaneous operations **only while on shift**. That is a product-specific rule, but it demonstrates why working time and concurrent capacity must be checked separately. [Infor, Infinite Scheduling resource field](https://docs.infor.com/csi/2026.10/en-us/csbiolh/inventory_user_cl_sl/mergedprojects/sl_invprod/fields/i/infinite_scheduling_resources.html)

Specify the scope of the relaxation: a resource, resource group, operation, planning horizon or entire run. State whether setup, labor, tooling, material availability and route precedence remain active. A system that relaxes machine loading need not relax every other constraint, and another product may use different settings.

Keep these questions separate:

| Planning question | Evidence required |
| --- | --- |
| When would an operation need to run? | Required dates, route, durations and calendar |
| How much capacity would that demand? | Resource demand by operation and time period |
| Can all operations actually run then? | Concurrent reservations, qualified resources and other constraints |
| May an order be released or a date promised? | The organization's release and commitment decision |

The last decision does not follow automatically from calculating a date.

## Loading and scheduling are related outputs

Oracle NetSuite's infinite-capacity documentation says the method does not account for current work-centre capacity or resource use and can therefore create overloads. Treat that as the documented behavior of the named scheduling method, rather than assuming every NetSuite configuration or scheduling feature behaves identically. [Oracle NetSuite, Infinite Capacity Scheduling](https://docs.oracle.com/en/cloud/saas/netsuite/ns-online-help/bridgehead_1508338236.html)

Conversely, a scheduler can generate dates without furnishing a load report. Microsoft's **Planning Optimization infinite-capacity feature** supports route sequences, parallel route structures, resource requirements, setup time and run time. Its documented limitations include no resource-load functionality and no route-scrap consideration. Do not infer that its scheduled dates include a complete capacity or yield analysis. [Microsoft, Scheduling with infinite capacity, updated October 3, 2025](https://learn.microsoft.com/en-us/dynamics365/supply-chain/master-planning/planning-optimization/infinite-capacity-planning)

For a capacity review, obtain or calculate the required load using a consistent resource and time basis. A resource-group total can hide a shortage on the only machine qualified for a particular operation. Likewise, a weekly total can conceal several jobs competing for the same hour.

## Original example: dates that conceal a daily overload

Assume an invented work centre with **one machine**, available from **08:00 to 14:00** on each of two days: **360 minutes/day**. Three jobs, A, B and C, each require **180 exclusive machine minutes**. All materials and operators are available at day 1's opening; there are no setups, interruptions, transfer delays or alternative machines. Each job is due at 14:00 on day 1.

An illustrative infinite forward schedule starts all three jobs at 08:00 and finishes each at 11:00 on day 1. Their route durations and due dates look acceptable individually, but the resource cannot run three exclusive operations simultaneously.

`day 1 required load = 3 × 180 = 540 machine minutes`

`load / available capacity = 540 / 360 = 150%`

The **180-minute shortfall** and **150% loading** describe required work relative to capacity. They are not actual utilization or permission to operate beyond a machine's capability. During 08:00–11:00, this hypothetical assignment requests three simultaneous machines even though only one exists.

One feasible sequence, using A–B–C priority, is:

| Job | Feasible assignment | Due-date result |
| --- | --- | --- |
| A | Day 1, 08:00–11:00 | On time |
| B | Day 1, 11:00–14:00 | On time |
| C | Day 2, 08:00–11:00 | Late |

Priority determines which job is late. To meet all original due dates, change the example’s capacity, durations, releases or commitments.

Two-day capacity is 720 minutes, exceeding the jobs' combined 540 minutes, yet that does not resolve the day-1 requirement. Capacity must be available when the operation needs it. A satisfactory aggregate total is insufficient evidence of an executable sequence.

## Resolve exceptions explicitly

Use the initial load profile to identify where assumptions need a decision. Candidate changes include authorized calendar extensions, qualified alternate resources, outsourcing, resequencing, lot splitting where allowed, or revised due dates. Evaluate their actual feasibility and costs before incorporating them. An unapproved overtime assumption is not available capacity.

After changing an assumption, recalculate both dates and resource demand. Check downstream operations, shared labor and fixtures, material timing and existing commitments. A local overload can move elsewhere; resolving machine time alone does not establish complete order feasibility.

ETH's classification connects capacity flexibility with due-date flexibility. Where neither can change enough, a planning calculation cannot manufacture a feasible solution. Infinite loading makes required work visible; it does not remove that conflict. [ETH Zürich, infinite and finite loading](https://opess.ethz.ch/course/section-5-3/5-3-4-overview-of-scheduling-and-capacity-management-techniques/)

## Preserve resource settings and planning provenance

Infor documents an interaction between resource-level and group-level infinite settings: a resource explicitly marked infinite remains infinite even within a non-infinite group. Other combinations depend on group membership. Review the effective setting rather than reading only the group's name. This is specific to its Scheduler, not a universal ERP precedence rule. [Infor CSI 2026.10, resource and group capacity interaction](https://docs.infor.com/csi/2026.10/en-us/csbiolh/inventory_user_cl_sl/mergedprojects/sl_invprod/fields/i/infinite_scheduling_resources.html)

For a handoff, retain the run date, scheduling direction, required dates, route/revision, quantity and duration basis, calendars, finite/infinite scope, resource qualifications and open capacity exceptions. Label requested, unconstrained calculated and capacity-feasible dates separately. Record which assumptions changed before treating an updated date as a production commitment.

## Source versions and limits

Reviewed **2026-11-01**. Microsoft's page identifies **October 3, 2025** as its update date. Infor's link belongs to **CSI 2026.10** documentation; that is a documentation version, not an independently verified page publication date. No separate issue date was established for the ETH or Oracle pages. Product behavior above is bounded to the cited feature and documentation; the example and handoff recommendations are original planning guidance.

## Related reference

[Production planning guides](index.md)