Inhalt
Key points
- An APS system computes realistic production plans from capacity, material, staff and constraints. The work shifts from building the plan to deciding on the basis of computed consequences.
- Production planning, production control and order management change most: they get answers where they previously had to estimate.
- Disruptions such as sick notes, material delays or line failures are assessed within minutes as a scenario with two to four options.
- The frozen zone keeps the short-term plan stable, and detailed scheduling and workforce scheduling interlock on a daily cycle.
- The decision stays with people – the APS system supplies the assessment of consequences, not the responsibility.
What does an APS system change fundamentally?
An APS system changes production planning fundamentally because it does not merely record plan versions but computes feasible plans and alternatives within minutes. The work therefore shifts from building a plan by hand to assessing scenarios and conflicting objectives.
The core of the change in one sentence: the plan is no longer built but assessed. The decision is between several feasible plans – and for the first time it rests on computed consequences rather than on experience and instinct.
For experienced planners this is at first an imposition: someone who has built good plans in their head for twenty years loses a recognised competence, and what replaces it is more demanding but less visible. Addressing this change of role openly decides acceptance more than any feature of the software.
The Aachen PPS model of the Institute for Industrial Management (FIR) at RWTH Aachen University describes exactly this structure: it explicitly assigns planning tasks to task holders, that is to positions within the company. Introducing an APS therefore changes not only functions but shifts tasks between positions – anyone looking only at the software misses the larger part.
Which data flows into an APS system – and what does each role get back?
An APS system only works reliably if it brings together data from sales, purchasing, time recording, the shop floor and maintenance: orders and requested dates, confirmed delivery dates and stock, attendance and qualifications, actual times and breakdowns, maintenance windows. The added value comes from every role getting back exactly the information it needs for its decisions – production a sequence per line, order management reliable dates, the warehouse staging dates, workforce scheduling the demand per qualification and management utilisation and bottlenecks.
Which roles does an APS system change?
An APS system mainly changes the work of production planning, production control, order management and workforce scheduling. Manual coordination becomes a decision on a shared data basis, and estimates become reliable planning.
| Role | Previously | With an APS and workforce scheduling |
|---|---|---|
| Production planning | Building the plan by hand, hunting for conflicts, maintaining lists, recalculating dates | Assessing and releasing the plan, computing scenarios, handling exceptions, maintaining the model and the rule set |
| Production control and supervisors | Translating instructions into a workable sequence each morning, assigning staff, improvising | Implementing the sequence, reporting deviations, leading and training instead of re-planning |
| Order management and sales support | Quoting a lead time from standard values or by asking production | Checking the date against the current plan and committing reliably; revising concretely when disruptions occur |
| Production engineering | Creating routings, with hardly any time for revision afterwards | Checking planned times against feedback, adding alternatives, modelling constraints |
| Purchasing and materials planning | Treating all dates alike, judging criticality from experience | Sees which procurement dates actually lie on the critical path and where float exists |
| Warehouse and internal logistics | Staging material on request or from a requirements list | Knows staging dates per line and the picking sequence in advance |
| Workforce scheduling | Shift roster separate from the production plan, reconciled in someone’s head each morning | Receives demand per qualification and time window; assigns people instead of hunting for gaps |
| Maintenance | Negotiating maintenance windows with production | Sees when a window costs least and can schedule it with justification |
| Production management | Overview through enquiries and daily meetings | Sees utilisation, bottlenecks and backlog at any time; decides on extra shifts with figures |
| Executive management and controlling | Metrics in retrospect from reporting | A forward view of utilisation, on-time delivery and capacity requirements for the coming weeks |
The common pattern
every role has to search and estimate less and decides faster on a reliable data basis. The time gained is the smaller effect – the larger one is that discussions start with the decision instead of with an argument about the facts.
Which processes change with an APS system?
An APS system changes not only the plan itself but also the processes around production planning, delivery commitments, material staging and staffing. Planning decisions are made faster, more reliably and more consistently.
From building the plan to the planning run
The plan is produced in minutes and can therefore be recomputed daily. That removes any reason to stick to an outdated plan – planning changes from an event into an ongoing state.
From status meeting to decision meeting
With a shared data basis, the part of the meeting that went into agreeing on the facts disappears. Meetings become shorter and less frequent but more demanding in substance – they are about trade-offs rather than reconciliation.
From standard lead time to a checked commitment
Delivery dates are checked against the current plan instead of being taken from a stored lead time – a capacity-checked commitment known as capable to promise. Sales also sees what an earlier date would cost.
From requirements list to staging date
The warehouse learns not only what is needed but when and at which line. Picking is then organised by sequence rather than by order number – an effect that is regularly underestimated.
From names to roles
Workforce scheduling computes with skill groups in the medium term and assigns specific people only at short notice. This increases plan stability and defuses co-determination and data protection questions. In detail: staff and skills in detailed production scheduling.
From escalation by shouting to a rule
How setup effort, on-time delivery, inventory and utilisation are weighed against each other is defined once and applied in every planning run – at MCP through the weighting rule. Renegotiating the same question every day comes to an end.
What does a working day with an APS system look like?
A working day with an APS system starts with importing the latest feedback, followed by a new planning run and the release of the sequence before the shift starts. New orders, disruptions and staffing issues are then assessed continuously against the current plan.
- Import the feedbackWhat is finished, what is running, where does the actual status deviate? Striking deviations point to wrong planned times and are noted, not merely corrected.
- Assess the planning runThe planner does not adopt the result blindly but checks conflicts and bottleneck resources and decides where to intervene.
- Release and distributeSequence to the lines, staging dates to the warehouse, staff requirements to workforce scheduling – from now on everyone works with the same version.
- Load new ordersDate enquiries are checked against the plan; for requested dates it becomes visible what they cost – usually moving two other orders.
- Handle unplanned eventsSick note, material delay, breakdown, rush order: who is affected, what options are there, what does each cost?
- Look ahead and check staffingWhere do bottlenecks arise next week, which qualification is missing, where is an extra shift worthwhile – in good time rather than on a Friday afternoon.
Why should detailed scheduling and workforce scheduling be connected?
An APS system is particularly effective in practice when detailed scheduling and workforce scheduling are connected. Only then can you judge whether an order is feasible not only in terms of equipment but also in terms of staff. The two exchange information every day:
- Available capacity flows inWorkforce scheduling supplies the capacity available per qualification and time window – including holidays, training and this morning’s sick note.
- Demand flows backFrom this the planning run produces the staff demand per operation. Workforce scheduling assigns people instead of hunting for gaps.
- Conflicts become visible earlyIf a qualification is missing, the run reports it for the coming week – not on the morning of the shift concerned.
The decisive convention: outside the frozen horizon, detailed scheduling computes with roles; inside it, workforce scheduling assigns names. If someone is absent, the assignment changes, not the sequence.
The most common mistake
running the production plan and staffing separately and bringing them together in the supervisor’s head each morning. That is not a reliable process but a dependence on one individual.
What a maintainable skills matrix looks like and what to observe on co-determination and data protection is covered in the article staff and skills in detailed production scheduling; the implementation is described by MCP Workforce Management.
What is the frozen zone in an APS system?
The frozen zone – also called a frozen period or time fence – is the short-term part of an APS system’s plan in which the production sequence is no longer changed freely. It protects picked material, setup states, staff assignments and commitments already made from unnecessary disruption; only defined interventions are allowed, for instance after a line failure or a material block.
Every change in this range creates real cost: returning material to the warehouse, setting up again, rescheduling shifts. Yet a system that recomputes in minutes tempts people to replan constantly – and that destroys trust on the shop floor faster than any bad plan. A graduated rule has proved its worth:
| Zone | Typical horizon | What may still be changed |
|---|---|---|
| Frozen | current shift to 1 day | Only after a breakdown or a material block; every change is logged and has to be justified |
| Stable | 2 to 5 days | Sequence within the setup family, assignment of people, rush orders after approval |
| Free | from about one week | Complete replanning with every run |
The exact length depends on the type of manufacturing: where setup takes long or batches are starting up, the frozen zone is longer; in assembly with short operations it is considerably shorter.
How does an APS system help with disruptions, failures and rush orders?
When something unplanned happens, an APS system shows within minutes which orders are affected and which options are realistic – for sick notes, material delays, line failures or rush orders. The events do not become rarer, but the answer is available in minutes rather than hours and can be justified.
| Event | The real question | What the system supplies |
|---|---|---|
| An employee calls in sick | Which operations are now unstaffed, and who could take them on? | The operations affected, a list of qualified alternatives with availability, the effect of postponing |
| Material arrives later | Which orders depend on it, and what do we bring forward? | All dependent operations, the new attainable date, sensible candidates to bring forward into the capacity released |
| A line fails | What does each route back into the plan cost? | Two to four options with their consequences for on-time delivery, cost and setup effort |
| A rush order is to be brought forward | Who pays for it – and with what? | The specific orders affected, their new dates and the additional setup effort |
| A customer asks about the delivery date | When will it really be finished? | A date checked against the current plan rather than a standard lead time or an enquiry to production |
| A customer wants it earlier | Is that possible, and at what price? | A feasibility statement showing which other commitments would have to give way |
| A quality problem, a batch has to be repeated | Where do we fit it in without pulling the rest apart? | A loading proposal taking cleaning transitions and material availability into account |
| A tool or fixture is broken | Which orders come to a standstill, and is there an alternative? | The operations affected across all orders, the permissible alternative resources |
| A qualification is understaffed | When does it become critical, and whom should we train? | A forecast of bottlenecks per qualification, a list of activities with only one available person |
| Maintenance needs a window | When does it interfere least? | A comparison of several windows with their effects on dates and utilisation |
The real difference
Before, the answer to these questions amounts to: “I’ll look into it and get back to you.” Afterwards: “Two orders move by three days, the third holds, and it costs a Saturday shift.” The same person, the same problem – only with figures instead of promises.
What benefits does an APS system bring to each department?
An APS system does not bring the same benefit to everyone involved, but it gives every role a concrete information advantage: production, order management, sales, purchasing, logistics, workforce planning and management each get the view they need for faster and more reliable decisions.
| Stakeholder | Specific result | What they can do with it |
|---|---|---|
| Production, per area and line | A justified sequence with start and end times, setup order and the qualification assigned | Work through instead of deciding; recognise deviations because there is a target state |
| Order management | A reliable date per order, and after a disruption a new date with a justification | Inform customers without asking around and without a safety margin |
| Sales | A feasibility statement on requested dates including the price in the form of orders moved | Negotiate knowing the consequence instead of committing on the spot |
| Warehouse and logistics | A staging date per item, line and time, plus the picking sequence | Pick ahead instead of reacting to requests; combine journeys |
| Purchasing and materials planning | Criticality per procurement item, the effect of date changes | Chase the supplier where it really counts |
| Workforce scheduling | Demand per qualification, time window and workplace | Assign instead of hunting for gaps; derive the training requirement |
| Maintenance | Assessed maintenance windows | Plan preventive maintenance instead of squeezing it between orders |
| Production management | An overall view of areas, utilisation, bottlenecks, backlog and on-time delivery | Decide on extra shifts and subcontracting in good time and with justification |
| Executive management and controlling | Forward-looking metrics instead of retrospect: expected utilisation, on-time delivery, capacity requirement | Take investment and staffing decisions on the basis of a computed outlook |
| Quality | A forecast of testing and release dates | Plan laboratory capacity instead of waiting for samples |
What changes with an APS system across several sites?
With several plants, the additional question is how planning, data model and rules can be standardised across sites. What matters is a shared core with local extensions – for the group with fifteen plants just as much as for the mid-sized company with a second site across the border.
The economic advantage lies less in the licence price than in reuse: the second site starts with a proven data model, finished interfaces and an objective weighting that has already been decided; what remains are the local constraints and master data. Different types of manufacturing are no obstacle, as long as the core model knows operations, resources, secondary resources and transition conditions. On the IT side, three points are decisive: a single release level, a clean separation of core and configuration, and a defined fallback mode in case of outages.
The governance question decides the rollout
If it is not settled who decides on changes to the shared core, the sites drift apart within a year – with the maintenance cost of five separate systems and none of the advantages. A good first site is one that is representative but not the most difficult.
Does working with an APS system differ by industry?
Yes, in its emphasis: an APS system follows the same planning principles across industries, but depending on the type of manufacturing, different constraints come to the fore.
Process industry and food
Planning works in blocks rather than individual orders: which campaign runs for how long, which transition comes next? Vessel occupancy and shelf lives constrain more than capacity does. In depth: campaign formation and cleaning optimisation and silo and tank scheduling.
Engineer-to-order and make-to-order
The focus is on order networks, procurement dates and completeness before assembly starts; the most frequent question is “what depends on this part?”. In depth: production planning in machinery and plant engineering.
Electronics manufacturing
Setup families and feeder allocation determine the day; whether a rush order breaks a family is decided several times a day. In depth: SMD line planning and setup optimisation.
Pharmaceuticals and medical technology
Release times, validated cleaning and documentation requirements make the plan less agile. That makes the outlook all the more important: what is not started today will not be released in two weeks.
What does an APS system not change – and where can it fail?
An APS system does not replace responsibility; it improves the basis for decisions. Decisions stay with people, and without maintained master data and clear rules even a good planning system loses its effect – it only makes a lack of discipline visible faster. The three most common pitfalls:
- Constant replanning without a frozen zone. If the sequence changes three times a day, the shop floor stops reading it.
- Manual interventions without feedback. Every recurring intervention is a rule missing from the model. If it is just made rather than reported, the model stays permanently worse than necessary.
- The plan as a suggestion. If every department still decides for itself whether to follow it, no shared data basis emerges – and the most important effect fails to materialise.
Maturity check: how far along is your production planning?
This self-assessment shows how far your production planning already works the way an APS system makes possible. Every statement that applies is a building block that already holds in daily work.
For manufacturers
Where is the biggest lever in your planning?
Every production operation plans differently. Let’s look at your process together: where do bottlenecks occur today, which use case offers the greatest leverage, and what results can realistically be achieved?
Book a potential analysisFrequently asked questions on working with an APS system
Yes, and that is usually the better route. A proven approach is to start with a bottleneck area or a representative plant, first define the frozen zone, role model and objective weighting there, and check the result against a historical reference week. Further areas, workforce scheduling or additional sites follow on this basis instead of changing everything at once.
At its core: orders with dates, routings with realistic operation and setup times, resources with calendars, material and stock data, and feedback from production; for workforce scheduling, qualifications and attendance are added. What matters is less the volume than the quality – planned times from costing lead to wrong plans even in the best system. The individual data objects are described in the article data objects for production planning.
Yes, with a changed and usually more demanding task: assessing scenarios, deciding on conflicting objectives, maintaining the rules and handling exceptions. Systems meant to replace the planner regularly fail on acceptance. Systems that relieve the planner of mechanical work succeed.
This requires a fallback mode that is defined and practised before go-live: an exported plan with sufficient range, clear responsibility for decisions during the outage and a procedure for catching up on feedback afterwards. A plant that improvises for twelve hours without a plan loses more than any optimisation gains – so the question belongs in system selection.
Yes, provided the core model supports them. Series, make-to-order and process production differ in the type of constraint, not in the type of planning: it is always about sequence, resource loading and dates. A core model with operations, resources, secondary resources and transition conditions covers all three; a system built for one type of manufacturing breaks down at the second – often only becoming apparent during the rollout.
Sources and notes
The description of roles, processes and the daily routine draws on project experience. The underlying framework of tasks, processes and task holders comes from the literature below. The figures in Figure 3 are an example, not a measurement.
- Aachen PPS model, developed at the Institute for Industrial Management (FIR) at RWTH Aachen University. It distinguishes the reference views of tasks, processes, functions and data and explicitly assigns tasks to task holders, that is to positions within the company. Presented among others in: H. Luczak, W. Eversheim (eds.), Produktionsplanung und -steuerung: Grundlagen, Gestaltung und Konzepte (production planning and control: principles, design and concepts), Springer.
- G. Schuh, V. Stich (eds.): Produktionsplanung und -steuerung 1 – Grundlagen der PPS (production planning and control 1 – principles). Springer. A current presentation of the reference model and the task view.
- H. Lödding: Verfahren der Fertigungssteuerung (Handbook of Manufacturing Control). Springer Vieweg, VDI series. Describes among other things which planned values production planning hands over to production control: input, output and sequence.
- VDI 5600 – manufacturing execution systems. Describes the task areas of an MES, among them detailed scheduling and control as well as personnel management.
- The terms frozen period, time fence and capable to promise are established in English-speaking practice and defined in the ASCM Supply Chain Dictionary.
Further reading: production planning metrics, the guide to production planning software and the overview of End-to-end planning.
Daniel Walkiewicz
Daniel Walkiewicz ist Partner bei MCP Algorithm Factory, verantwortet dort die Produktionsfeinplanung und APS-Prozessberatung und setzt seit 2010 Planungsprojekte bei MCP um. Das Unternehmen hat als Ganzes über 300 Planungsprojekte in Betrieb genommen und betreibt eines der größten Opcenter-APS-Teams in Europa.