Detailed planning in discrete manufacturing

The reality of discrete manufacturing is dynamic: customers change priorities, materials arrive late, machines break down, shifts are rescheduled. This is precisely why a ‘plan from yesterday’ rarely lasts until the end of the week. Production planning must be continuously adapted to shop floor realities. This is exactly where detailed planning comes into play.

What is an Advanced Planning & Scheduling System and why is it needed?

An Advanced Planning & Scheduling System (APS) addresses precisely this: it dynamically supports detailed, capacity-oriented planning of individual operations and ensures that orders are scheduled on the right machine at the right time, whilst all necessary conditions – such as the availability of materials, personnel and tools – are met.

It is important to understand the role of detailed planning within the overall planning process: sales planning and rough planning – typically carried out via the MRP run in the ERP system – define what is fundamentally required and in what quantities. Detailed planning answers the core operational question that builds on this: “When and where will specific orders actually be produced, taking real-world conditions into account?”.

This is precisely where an APS differs from many ERP planning logics. An APS takes finite capacities and real-world constraints (such as shift patterns, set-up times, alternative resources, tools and qualifications) into account, thereby generating actionable plans rather than theoretical dates.

How an Advanced Planning & Scheduling System coordinates production

At the heart of an APS lies the ability to calculate a consistent, capacity-feasible plan from a wealth of individual data points. In discrete manufacturing, this means: orders consist of operations that run on workstations, often involving alternative machines, sequences dependent on set-up times, and scarce ancillary resources such as tools and personnel.

An APS schedules these operations whilst taking actual resource constraints into account.

Typical operational decisions that are made (or at least prepared) in the APS are:

• When does which order start, down to the minute?
• On which machine or line does the order run?
• In what order should the orders be processed to minimise set-up times and waiting times?
• How do we respond to disruptions, changes in priority or material delays?

In discrete manufacturing in particular, sequence planning is often a powerful lever for improving key performance indicators.

An example: If set-up times are sequence-dependent (e.g. colour, material, tool change), the sequence determines whether a day runs smoothly – or whether the shift is ‘wasted’ on changeovers.

A good APS supports rapid recalculations as soon as boundary conditions change.

In this way, detailed planning directly influences key performance indicators such as on-time delivery, lead times, work-in-progress (WIP) and machine utilisation.

Integration of the Advanced Planning & Scheduling System into the system landscape

An APS delivers its greatest benefit when it does not run in isolation, but is embedded within the relevant systems for planning, controlling and executing production processes.

Ideally, an APS is part of a closed-loop system:

• The ERP provides the data foundation: orders, bills of materials, work plans, stock levels and key dates.
• The APS uses this to generate the detailed plan: sequence, start and end times accurate to the minute, resource allocation subject to constraints.
• The MES implements this in terms of execution: approvals, dispatching/work queue, confirmations, quality and status data.
• Shop floor feedback (e.g. via confirmations, stoppages, progress) flows back and updates the detailed planning and, ideally, the ERP schedule as well.

One aspect of integration that is often underestimated is the process rule: the MRP run must not constantly ‘plan into the detailed schedule’. To this end, time fences are used in many environments – defined time windows within which the plan should only be altered to a limited extent, as short-term changes generate disproportionate effort and knock-on effects. In this case, the motto is: stability in the short term, flexibility in the long term.

Six advantages of an Advanced Planning & Scheduling System

Companies rarely implement an APS simply because it sounds modern, but because operational pressure is mounting: smaller batch sizes, more variants, volatile supply chains, and a shortage of skilled workers. Without an APS, this quickly leads to high complexity in discrete manufacturing, with more manual coordination, greater re-planning effort and bottlenecks identified too late. Organisations end up reacting rather than acting proactively. Managing this complexity is one of the greatest advantages of an APS.

The following six examples stand out in particular:

Realistic deadlines instead of wishful ones

The APS plans against limited capacities (finite planning). This means that if there is no capacity available, the order is not ‘squeezed in somewhere’, but is neatly scheduled into the next available slot, and the delivery date can be reliably derived accordingly.

Higher productivity through better sequencing

An APS can create sequences in such a way as to reduce changeovers. In discrete processes with high changeover rates, this is often one of the most effective ways to boost productivity.

Shorter lead times and less work in progress

When the sequence is logical and bottlenecks become visible, there are fewer queues at critical workstations. This tends to reduce work in progress (WIP). The material flow becomes smoother, rather than a constant stop-and-go.

Less planning effort, more focus on exceptions

A good APS does not automate ‘planning as a thought process’; it automates routine calculations: capacity checks, sequence proposals, schedule reconciliation. As a result, planning becomes less about manual Excel administration and more about managing exceptions. Planners and production controllers are enabled to answer questions such as ‘What is critical today and what are we doing about it?’.

Transparency & alerts instead of surprises

Planning under constraints makes bottlenecks visible earlier. Additionally, deviations (e.g. downtime, missing materials) can be fed back into the plan as events, allowing consequences to be assessed more quickly. This also supports structured shop floor communication.

Order networks and pegging: reliably managing dependencies across BOM levels

Particularly in discrete manufacturing, orders rarely consist of “isolated islands” but rather of order networks: pre-assemblies, sub-assemblies and final assemblies are interdependent across multiple BOM levels. An APS is particularly valuable here because it not only takes dependencies into account in terms of scheduling but also makes them transparent.

The key term for this is pegging. Pegging describes the linking of requirements (for example, a sales order/production order) with suitable sources of supply (for example, available stock, derived production orders for secondary requirements or purchase order items) across all BOM levels. This creates a pegging structure – a network of interconnected orders that maps the flow of materials and orders from raw materials through assemblies to delivery.

Why is this an advantage in detailed planning? Because it means the APS does not merely ‘plan nicely’, but makes the consequences of changes far easier to manage:

• Transparency regarding what a part is actually needed for: If a critical part is missing, pegging shows which sales order or assembly is behind it – rather than just an anonymous material requirement figure.
• Better prioritisation in the event of bottlenecks: When components are in short supply, targeted decisions can be made regarding which order chain should be supplied first, rather than on a ‘first come, first served’ basis.
• Faster impact assessment: If a supplier part is delayed or a pre-assembly step is cancelled, it is immediately clear which downstream operations and delivery dates are affected.
• Synchronised planning of assemblies and final assembly: Network relationships ensure that sub-assemblies are available on time when final assembly requires them.

Instead of optimising many individual orders independently, the APS plans an order chain as a coherent network. This makes detailed planning not only more realistic but also more robust, because dependencies are visible and decisions (prioritisation, rescheduling, material allocation) are not based on assumptions.

Challenges in implementing an APS

However positive an APS may be, the benefits do not arise ‘simply by installation’, but through well-defined process rules and clear responsibilities. An APS has an impact on adjacent processes. Therefore, a process-based framework is crucial.

Interfaces must be defined not only technically, but also in terms of processes

A change in detailed planning can trigger knock-on effects: orders are rescheduled, delivery dates change, and order confirmations may be automatically regenerated, depending on the configuration of the ERP processes. For this reason, clear rules are needed: which planning results may be automatically written back to the ERP system, and which only after approval?

Time Fences & Firming Logic: Stability in the Near Term

Without a ‘protective fence’, production becomes unsettled: constantly changing sequences, constantly shifting priorities. Time-fence concepts create a stable immediate scope (for example, the next few days or shifts) and only allow adjustments outside this window. These must be precisely defined to strike a balance between an optimal plan and a calm production environment.

Full automation is rarely the end goal

In discrete manufacturing, there are bottlenecks that can be resolved organisationally at short notice (reassigning staff, extra shifts, temporary staff, alternative workstations). Therefore, a multi-stage approach to planning often makes sense, rather than creating a fixed plan from the outset. It may therefore make sense to first plan machines or lines (which are physically fixed) and then, in a second iterative step, carry out the staffing and skills matching (which can be flexibly arranged). The ‘degree of optimisation’ must therefore be defined depending on the complexity of production.

Prioritisation rules must be explicit

An APS requires clear rules: customer orders often need to be prioritised higher than production orders required to replenish safety stock. Orders to replenish safety stock, in turn, are often prioritised higher than orders to fulfil anonymous customer forecast requirements. Without such rules, the plan appears ‘incorrect’, even though it is merely executing the implicit assumptions provided to it via dates.

Handling scheduling logic and delayed orders correctly

If, from a logical perspective, an order is ‘already delayed’, a system may unintentionally keep pushing it further back to protect other deadlines. This can be sensible – or disastrous, if that very order is strategically important. Therefore, heuristics or optimisation objectives (e.g. on-time delivery vs. setup time minimisation vs. WIP) must be deliberately selected and tested so that the APS generates acceptable plans.

Data quality as a decisive factor

Shift calendars, setup matrices, actual durations, auxiliary resources (such as tools and fixtures), qualifications: If this information is missing or not maintained, an APS cannot demonstrate its full potential. Sufficient data quality is required to generate reliable plans. It should be noted here that data quality does not need to be perfect at the time of implementation. Thanks to the transparency provided by an APS, poor data quality becomes quickly apparent through unreliable plans and requires immediate action. This can be used as an opportunity to gradually improve data quality as part of an APS implementation.

Given the challenges described above, a pragmatic approach should be adopted for the successful implementation of an APS, following the principle of first creating a good and feasible plan, then an optimal one. This can be achieved by first (Stage 1) creating transparency using the APS and beginning to reschedule manually, then (Stage 2) defining the rules and utilising the automated processes of dynamic detailed planning, and finally (Stage 3) iteratively refining the planning by taking further conditions into account.

Conclusion

Detailed planning is the operational heart of discrete manufacturing: This is where strategic guidelines and rough planning meet the reality of machines, people, materials and tools. An APS makes this level planable, not theoretically, but in a capacity-realistic way.

This increases on-time delivery and stability, reduces lead times and inventory levels, and shifts the planning effort from manual coordination to structured control.

However, the benefits do not arise automatically. Crucial factors include seamless integration into the system landscape, clearly defined planning horizons, prioritisation and firming rules, as well as robust feedback structures. It is precisely then that planning becomes genuine control and the APS transforms from an IT tool into a strategic success factor.

Not every APS is equally well suited to every manufacturing operation. Systems differ in their strengths (e.g. variant/customised production vs. batch production, optimisers vs. heuristics, ancillary resources, multi-site, user interface). Those who clearly define and prioritise their own conflicting objectives (delivery reliability, lead times, stock levels, capacity utilisation) can select the right APS with far greater accuracy – and realise the benefits more quickly.