On-Time Delivery Case Study: When the Plan Doesn't Account for People
Case study of a mid-sized manufacturer of customer-specific industrial components: through a realignment of production and assembly planning, on-time delivery was raised from 50 % to 90 %.
The starting position in 30 seconds
Situation. The company is a mid-sized industrial business in the German-speaking region that produces customer-specific components for industrial applications. Assembly is carried out at specialized workstations. Series orders, call-offs from framework agreements and repairs are processed in parallel in the same assembly area.
Complication. Delivery dates were repeatedly postponed, and order backlogs increased. Planning was based on start dates rather than on the promised delivery date. Absences such as holidays were not taken into account in capacity planning, nor were the differing qualifications of employees. Missing parts were only identified after the order had been handed over to assembly.
Solution. First, on-time delivery was introduced as a KPI. Planning was then aligned with the delivery date: realistic master data in the ERP system, backward scheduling, capacity planning based on actual attendance, assignment by qualification, binding weekly planning and a material preview for the early identification of missing parts.
The project was part of a broader restructuring. This state is characterized by fewer problems being solved per unit of time than new ones arising. Postponed delivery dates cause additional effort – queries, replanning, rush orders – which in turn ties up capacity needed to meet further dates.
The starting point: introduction of the on-time delivery KPI
At the start of the project, on-time delivery was not systematically recorded. The KPI was therefore introduced, including a binding definition of when an order is considered on time. The first reliable value was 50 %; every second order was delivered late.
In the first months after its introduction, the value initially declined, as previously unrecorded backlogs became visible. With this data basis, however, the effectiveness of the subsequent measures could be assessed for the first time.
Lever 1: Reliable master data
The quality of planning depends on the quality of the stored times. Routings were created for all products, pre-assembly and final assembly were defined as separate, plannable operations, and external processing steps such as surface treatment were modelled. Replenishment and throughput times were maintained realistically; standard queue times were defined for kitting, assembly and testing. Since then, the ERP system has delivered dates that can be met in assembly.
Lever 2: Backward scheduling from the delivery date
Scheduling was converted from forward to backward logic: the starting point is the delivery date expected by the customer, from which the dates for kitting, assembly and testing are derived.
In addition, capacity is planned on the basis of actual attendance, taking holidays and other absences into account. Orders are assigned by qualification, based on a skills matrix that also identifies training needs.
Lever 3: Binding weekly planning
Based on backward scheduling, a binding order plan per workstation is drawn up each week for the following week. Before release, work preparation checks for missing parts. Orders are kitted in the preceding week, so that they are handed over to assembly with complete material. Remaining missing parts are escalated, and purchasing reports back a confirmed delivery date. An order board in assembly visualizes the weekly plan.
Lever 4: Early identification of missing parts
A material preview fully explodes the bills of material of upcoming orders and matches requirements against stock and open purchase orders. The ordering horizon was extended considerably so that parts with long replenishment times are scheduled in good time. Overdue supplier dates are monitored and followed up as a separate KPI.

Steering: daily and weekly reporting
To embed the changes, planning was integrated into management routines. A forward-looking KPI report shows on-time delivery daily and weekly, the target/actual comparison of delivered orders against the weekly plan, and order backlogs – broken down into self-inflicted and external causes. Deviations are thus identified early and addressed in the next coordination meeting.
The results
- On-time delivery raised from 50 % to 90 %.
- Reduction in self-inflicted order backlogs since separate monitoring was introduced.
- Binding weekly plans instead of constant replanning in assembly.
- Identification of missing parts before assembly starts.
Progress was not linear. The target of 95 % had not yet been reached by the end of the period under review. The remaining gap depends primarily on suppliers' delivery reliability.
Three key takeaways
- Measurability as a prerequisite. A clearly defined on-time delivery KPI is the basis for any improvement, even if the starting value is unfavourable.
- Planning based on real capacity. If attendance and qualifications are not taken into account, capacity that is not available is scheduled and dates are promised that cannot be met.
- On-time delivery is determined in work preparation. Master data, backward scheduling and material preview determine adherence to delivery dates before an order reaches assembly.
Note: This case study is based on a real 10-P Consult engagement. The client's name and identifying details have been anonymized or generalized.
Further articles:
Operational Excellence Case Study: When Assembly Spends More Time Searching Than Assembling (link)
Defining KPIs: How a Business Goal Becomes a KPI That Management Trusts (link)
Peak Performance through Transparency (Part 1 / 2) (link)
Where does your company stand?
Is your company facing similar challenges? The Quick Health Check takes five minutes to complete and shows which state your company is in and which fields of action offer the greatest leverage – as a five-page profile, reviewed by an experienced consultant and free of charge.
Start the Quick Health Check

