Configuring Supply Planning Master Data, Key Figures, and Time Series for Constrained Planning
Learn how to configure the master data, key figures, and time-series structures required to move from a basic unconstrained supply view to a constrained supply planning setup that respects capacity and sourcing rules.
Explanation
Moving from conceptual understanding to a working constrained supply plan requires deliberate configuration of several interdependent layers: master data, key figures, time profiles, and the run profile that ties them together. This lesson focuses on what an intermediate consultant configures and validates before a constrained heuristic or optimizer run can produce trustworthy results. Master data first. Location Product records must carry attributes such as lot sizing rules, safety stock parameters, and shelf life if relevant, because these directly influence how supply proposals are rounded, batched, and timed. Production Source and Transportation Lane (or equivalent sourcing) records define which plants can produce which products and which lanes can move them between locations, along with lead times and, where applicable, cost fields used by the optimizer. Resource master data defines available capacity per time bucket (for example, hours per day per line), and Production Process Model or Production Data Structure elements define resource consumption rates per unit produced. Any gap here—such as a missing transportation lane or an incomplete resource calendar—will either cause the run to silently ignore a valid sourcing option or, worse, produce infeasible or unrealistic results that look plausible. Key figures come next. A constrained run typically needs, at minimum: a demand input key figure (often the consensus or constrained demand from an upstream process step), a supply proposal key figure (often split by production, procurement, and transportation), a capacity consumption key figure linked to resources, and inventory-related key figures such as projected stock and safety stock. Many implementations also configure exception-driven key figures, such as late confirmation or unmet demand, to help planners quickly identify infeasibilities after a run. Key figures must be correctly typed (stored vs. calculated), assigned to the right aggregation/disaggregation behavior, and placed at the correct planning level; a key figure defined at a coarser level than the master data it should reflect will produce misleading aggregated numbers. Time profiles and planning horizons matter significantly in Supply Planning because capacity and lead times are inherently time-phased. Consultants must configure the planning horizon (how far into the future the run considers), the time bucket (day, week, month), and often a separate frozen or firmed horizon inside which the system should not change already-committed supply proposals. Getting the frozen horizon wrong is a frequent source of planner distrust, since a run that keeps changing near-term proposals every cycle undermines confidence in the process. Run profile configuration ties these elements together: it specifies whether the run is unconstrained heuristic, constrained heuristic (which respects capacity but uses priority-based logic rather than true optimization), or optimizer-based (which can minimize cost or maximize service subject to constraints, using linear or mixed-integer programming). The choice affects both the required master data completeness and expected run time; optimizer runs are more sensitive to data quality issues such as missing costs or inconsistent unit-of-measure conversions, because infeasibilities can cause the solver to fail or return degenerate solutions. Validation before go-live should include running the heuristic first to sanity-check basic supply flow, then progressively enabling constraints and comparing proposal quantities against known capacity limits, and finally reviewing exception key figures for unexpected values. Differences across SAP IBP releases and template versions (for example, availability of certain optimizer cost key figures) mean this configuration should always be verified against the specific tenant rather than assumed identical to another project's setup.
Real project scenario
An automotive parts supplier is extending its IBP model from unconstrained supply planning to a constrained heuristic run because capacity at a bottleneck resource is repeatedly exceeded in initial pilot results. The consulting team audits Location Product and Production Process Model records and discovers several products are missing resource consumption rates, causing the heuristic to treat those products as having zero capacity impact. After correcting the master data and re-running, the exception key figures for late confirmation drop significantly, and the business gains confidence in the constrained proposals before extending the approach to additional plants.
Common mistakes
• Enabling a constrained or optimizer run before verifying that all relevant Production Sources, Transportation Lanes, and resource capacities are complete. • Defining key figures at a planning level that does not match the granularity of the underlying master data, causing misleading aggregation. • Ignoring frozen or firmed horizon settings, resulting in near-term supply proposals that change unpredictably between runs and erode planner trust. • Assuming the optimizer will simply 'work around' missing cost or capacity data instead of producing infeasible or misleading results. • Skipping incremental validation (heuristic first, then constrained heuristic, then optimizer) and jumping straight to the most complex run type.
Best practices
• Complete and validate master data (sourcing, resources, lead times) incrementally before enabling capacity constraints. • Configure frozen/firmed horizons deliberately in partnership with planners to balance stability and responsiveness. • Build exception key figures early so infeasibilities and unmet demand are visible immediately after each run. • Validate constrained results against known real-world capacity limits before trusting the run for planning decisions. • Progress run complexity incrementally: unconstrained, then constrained heuristic, then optimizer, validating at each stage.
Interview angle
Interview questions at this level often ask candidates to explain the practical difference between constrained heuristic and optimizer runs, and what master data gaps typically cause infeasible or unrealistic results. Strong answers connect specific master data objects (Production Source, resource capacity) to specific run behaviors rather than giving only textbook definitions.