The Operating Model for Manufacturing Systems

Changeovers and contamination rules often shape more of the day than the nominal cycle time does. A line that changes families ten times will spend significant minutes in setup even if each change is short. If the process runs in a cleanroom, windows, gowning, and air classifications limit when and where you can run, which further reduces freedom. The outcome is late orders and weekend firefighting. Two simple disciplines change this. First, reduce the work inside each change with SMED so more of it happens while the line is running. Second, model sequence-dependent rules and cleanroom windows in Opcenter APS so the schedule minimizes setup and respects regulated constraints automatically (Shingo, 1985; European Commission, 2022).

Start with data that matters. You do not need a study of every tool to begin. Collect three items for the main constraint and its feeders: setup families with clear names, average internal setup minutes by family-to-family change, and any cleanroom or campaign rules that apply. Family names should be meaningful on the floor, such as color code, chemistry, alloy, flux, nozzle set, or sterile class. Opcenter APS uses a setup matrix that maps from-family to-family times so the solver can trade sequence against due dates. The same matrix also helps your SMED effort because it shows which changes are most expensive and therefore deserve a kaizen first (Siemens Digital Industries Software, n.d.; Shingo, 1985).

Define setup families with care. A family is a group of jobs that can run back to back with very little internal setup. The right family reduces cleaning, tool change, or requalification. Academic reviews show that sequence-dependent setup scheduling is a major lever in flow shops and job shops, with significant impact on lateness and throughput when families are chosen well (Allahverdi et al., 2008). In Opcenter APS, enter a short code for each family and maintain the matrix centrally. If your process uses tools or fixtures, treat those as secondary constraints with limited counts so the schedule knows when a tool is not available, not only when a machine is free (Pinedo, 2016; Siemens Digital Industries Software, n.d.).

Build campaign logic for flows that prefer long runs. Many clean or sterile processes benefit from campaigns that limit the number of openings or clean-down events. Campaigns can be modeled as time windows or minimum run lengths. For example, a sterile filling line may operate within a Class ISO 5 zone nested in an ISO 7 room. Annex 1 requires a documented contamination control strategy and verification that conditions are maintained, which is easier when the schedule enforces campaign windows and reduces breaks that require interventions (European Commission, 2022; ISO, 2015). In Opcenter APS, create a calendar event that represents a campaign window and a rule that penalizes or forbids mid-window family changes. Then preview two schedules with and without the window so everyone sees the trade.

Translate cleanroom rules into schedulable constraints. ISO 14644-1 defines air cleanliness by particle concentration, and ISO 14644-5 provides operational guidance such as access control and gowning that influence practical capacity (ISO, 2015; ISO, 2004). If only certain rooms can host a sterile class, mark those resources as eligible for specific product families. If a room requires a recovery time after an intervention, add a cooling or recovery setup that the matrix applies when operators must break containment. If gowning limits crew mobility, add crews as secondary resources with calendars so the schedule does not over-commit the same people to multiple rooms at once (Pinedo, 2016; Siemens Digital Industries Software, n.d.).

Run an Ask this → Get that loop to make progress in one week.

1. Ask: What is the single most expensive family change on the bottleneck. Get: a targeted SMED and sequencing play. Reduce internal steps, convert what you can to external, standardize tools, and then increase batch size or sequence so that change appears fewer times per day. Academic and practitioner literature reports 50 to 90 percent internal setup reduction when SMED is applied with discipline, which compounds when paired with sequence-dependent scheduling (Shingo, 1985; Allahverdi et al., 2008). Screenshot idea: a side-by-side schedule before and after the matrix is applied. Alt text: “Two Gantt charts for the bottleneck, showing fewer family changes and lower setup time after matrix rules.”
2. Ask: Which cleanroom rules create the most misses or rework. Get: codified campaign windows and recovery setups. Annex 1 requires control and verification rather than improvisation. Encoding windows in APS prevents a class violation from appearing as a surprise on the day of production (European Commission, 2022; ISO, 2015). Screenshot idea: a calendar overlay that highlights allowed windows. Alt text: “Schedule view with green campaign window blocks and jobs aligned inside them.”
3. Ask: What secondary constraints are always the excuse. Get: an explicit model for tools, fixtures, and crews. When a tool is quantified, the schedule can stop double-booking it. When a crew is modeled, the plan can stop assigning them to incompatible rooms at the same time. This keeps the morning from starting with manual shuffling (Pinedo, 2016; Siemens Digital Industries Software, n.d.).

Prove the value with a simple experiment. Take last week’s orders for one family. Generate a schedule with no setup matrix and then with the matrix enabled. Record total setup minutes, late jobs, and cleanroom window violations. Teams regularly see double-digit setup reduction with no capital, which aligns with the literature that shows how sequence-aware schedules reduce both setup and tardiness when setup times are a large share of the cycle (Allahverdi et al., 2008; Pinedo, 2016). If you pair the change with a short SMED exercise, you can often remove one change per shift, which is a win that operators feel immediately (Shingo, 1985).

Connect scheduling to execution so gains stick. Publish the schedule each day to Opcenter Execution or to the dispatch list that operators already use. When an unplanned cleaning or hold occurs, record the event with a reason so the matrix and windows can be tuned. If a job misses a window, review the cause and either improve SMED or adjust the calendar so the next run fits. This feedback loop keeps cleanroom rules as part of the normal plan rather than as a last-minute constraint that forces overtime (Siemens Digital Industries Software, n.d.).

Watch a short KPI set that proves improvement. Track changeover minutes per shift, percent of jobs inside cleanroom windows, and on-time starts at the bottleneck. Add a small accessibility-friendly dashboard that labels lines directly and includes alt text like, “Line chart of setup minutes per shift with goal line and last four weeks of results.” Simple views make it easy to share progress in daily and weekly forums without specialty tools (W3C, 2023).

Close with a 90-day plan. In month one, create the first setup matrix and cleanroom windows for one line. In month two, run two SMED events on the worst changes and model crews or tools as secondary constraints. In month three, expand families to a second line and publish a schedule that covers both, then share the win in a short story with photos of tool carts or standardized kits. These are the kinds of improvements that your planners and operators will sustain because they make the day calmer and safer (World Economic Forum, 2025).

Mini FAQ 

References

• Allahverdi, A., Ng, C. T., Cheng, T. C. E., & Kovalyov, M. Y. (2008). A survey of scheduling problems with setup times or costs. European Journal of Operational Research, 187(3), 985–1032. https://doi.org/10.1016/j.ejor.2006.06.060
  This survey is relevant because it compiles results on sequence-dependent setups across common shop environments. It covers models, objectives, and heuristics that show why setup-aware sequencing reduces tardiness and total setup time when families are defined well. Two takeaways are that setup times change optimal sequences and that small, practical heuristics often capture most of the benefit.
  
  
• European Commission. (2022). EU GMP Annex 1: Manufacture of sterile medicinal products. https://health.ec.europa.eu/system/files/2022-08/20220825_annex1_en.pdf
  This annex is relevant because it sets expectations that drive cleanroom scheduling and campaign behavior in regulated plants. It covers contamination control strategy, qualification, environmental monitoring, and intervention management. Two takeaways are that schedules must protect cleanroom states and that verification evidence should match how work is sequenced.
  
  
• International Organization for Standardization. (2004). ISO 14644-5: Cleanrooms and associated controlled environments — Part 5: Operations. https://www.iso.org/standard/34232.html
  This standard is relevant because it provides operational practices that influence capacity and scheduling inside cleanrooms. It covers personnel flow, gowning, material movement, and maintenance that affect when runs can occur. Two takeaways are that operational constraints belong in the plan and that stable practices reduce unplanned cleaning.
  
  
• International Organization for Standardization. (2015). ISO 14644-1: Cleanrooms and associated controlled environments — Part 1: Classification of air cleanliness by particle concentration. https://www.iso.org/standard/53394.html
  This standard is relevant because cleanroom classification defines the conditions your schedule must maintain. It covers particle concentration limits, testing, and classification methods. Two takeaways are that class constraints can be modeled as resource eligibility and that recovery times after interventions must be planned.
  
  
• Pinedo, M. (2016). Scheduling: Theory, algorithms, and systems (5th ed.). Springer. https://link.springer.com/book/10.1007/978-3-319-26580-3
  This book is relevant because it explains sequence-dependent setup scheduling and secondary resource constraints that appear in real plants. It covers models and algorithms with practical heuristics that map to APS capabilities. Two takeaways are that setup matrices change optimal order and that adding crew or tool constraints makes schedules realistic.
  
  
• Shingo, S. (1985). A revolution in manufacturing: The SMED system. Productivity Press. https://www.routledge.com/A-Revolution-in-Manufacturing-The-SMED-System/Shingo/p/book/9780915299036
  This book is relevant because it is the primary reference for reducing internal setup and converting tasks to external work. It covers principles and step-by-step methods that deliver large setup reductions when applied with discipline. Two takeaways are that SMED and sequencing reinforce each other and that visual standards keep gains from fading.
  
  
• Siemens Digital Industries Software. (n.d.). Opcenter Advanced Planning and Scheduling (APS). https://plm.sw.siemens.com/en-US/opcenter/advanced-planning-scheduling-aps/
  This page is relevant because it documents APS capabilities that encode sequence-dependent setups, calendars, and secondary constraints. It covers planning and scheduling features with case evidence for productivity and due-date gains. Two takeaways are that modeling real rules frees capacity and that side-by-side scenarios help teams choose sequences they trust.
  
  
• University of Cambridge Institute for Manufacturing. (n.d.). Finite capacity scheduling. https://www.ifm.eng.cam.ac.uk/research/dstools/finite-capacity-scheduling/
  This primer is relevant because it clarifies how finite scheduling places operations on specific resources with limited capacity. It covers methods and objectives that fit high-mix manufacturing. Two takeaways are that finite models prevent overloads and that different heuristics can reach similar results when fed with good setup data.
  
  
• World Economic Forum. (2025). Global Lighthouse Network 2025: The mindset shifts driving scale. https://reports.weforum.org/docs/WEF_Global_Lighthouse_Network_2025.pdf
  This report is relevant because it shows how standardized data and short learning cycles sustain improvements such as setup reduction and stable schedules. It covers operating models and quantified results across sectors. Two takeaways are that daily cadence beats large projects and that shared definitions allow replication across sites.
  
  
• World Wide Web Consortium. (2023). Web Content Accessibility Guidelines (WCAG) 2.2. https://www.w3.org/TR/WCAG22/
  This guideline is relevant because schedule and KPI visuals must be readable by everyone. It covers text alternatives, contrast, and structure that help assistive technologies communicate context. Two takeaways are that concise alt text improves access and that color plus labels avoids ambiguity.
  
  
• Zhu, X., Bard, J. F., & Yu, G. (2001). Disruption management for resource-constrained project scheduling. Journal of the Operational Research Society, 52(7), 764–773. https://doi.org/10.1057/palgrave.jors.2601156
  This article is relevant because cleanroom events and unplanned changeovers create disruptions that schedules must absorb. It covers methods to revise schedules quickly while honoring resource constraints. Two takeaways are that fast rescheduling reduces lateness and that modeling constraints up front simplifies recovery.