CSCM common mistakes
CSCM · Understand the traps and how to avoid them.
Treating the four classical supply chain stages (plan, source, make, deliver) as strictly sequential and answering integration questions as if each stage owns its own KPIs, so questions asking how a demand signal propagates through the chain get answered stage by stage instead of end to end.
Study material often presents the stages as a numbered list, and candidates memorize the list order rather than the feedback loops between stages; exam items that hinge on cross-stage trade-offs then get answered with single-stage logic.
What to do instead
Before answering any strategy item, draw the loop: demand signal → plan → source/make → deliver → customer feedback. For each option, ask which stage bears the cost and which stage captures the benefit. If an option optimizes one stage while shifting cost downstream without naming the trade-off, it is usually the distractor.
Choosing the forecast method with the lowest error on the historical series without checking whether the series has trend or seasonality, so a simple moving average is selected for data that clearly trend upward.
Candidates memorize MAPE/MAD comparisons as a mechanical step and skip the pattern-diagnosis step; moving averages lag trend by construction, which produces deceptively acceptable-looking error on noisy data.
What to do instead
Run a fixed sequence before selecting a method: plot or describe the series, classify it (level only, trend, seasonal, intermittent), then match method to pattern - exponential smoothing for level, Holt for trend, Holt-Winters or decomposition for seasonality, Croston-type methods for intermittent demand. Only compare error metrics among methods appropriate to the diagnosed pattern.
Confusing forecast bias with forecast error, and reporting a low MAD as proof the forecast is unbiased when the errors cancel out in the average.
Both are 'accuracy' metrics, and candidates treat them as interchangeable. A forecast that consistently over-forecasts by a small amount can still show a modest MAD while cumulative bias accumulates inventory.
What to do instead
Check both: MAD/MAPE for dispersion, and running sum of forecast errors (bias) for direction. If bias is materially nonzero, the fix is model or process correction, not a tighter error band. On exam items, look for the word 'consistent' or 'systematic' - that signals bias, not random error.
Treating supplier selection as a lowest-price decision and ignoring total cost of ownership, capability, and risk, so questions about sourcing strategy get answered with the cheapest quote.
Price is the most visible and easily compared number, and candidates underweight qualitative factors such as quality history, financial health, and capacity because they are harder to score.
What to do instead
Build a weighted scorecard in your head: price, quality, delivery reliability, capacity, financial stability, geographic/geopolitical risk, sustainability compliance. For each answer option, ask whether it addresses only price or the full cost and risk picture. Options that name a single criterion as decisive are usually wrong.
Mixing up the roles of S&OP and master production scheduling, answering S&OP questions with MPS-level detail (specific SKUs, weekly buckets) or MPS questions with aggregate family-level language.
Both are planning processes and both produce 'a plan', so candidates blur the time fence and aggregation level. S&OP works at product-family level over monthly horizons; MPS works at end-item level over weekly/daily buckets within the frozen zone.
What to do instead
Anchor on two axes: aggregation (family vs end item) and horizon (months vs weeks). If the stem mentions families, consensus, or demand-supply balance, it is S&OP. If it mentions specific items, due dates, or the frozen zone, it is MPS. Reject any option that violates the time fence or mixes the two levels.
Applying a single EOQ or reorder-point formula without checking whether demand and lead time are deterministic or variable, so safety stock is set to zero in a stochastic environment.
Textbook formulas are memorized in their simplest deterministic form, and candidates do not notice that the stem describes variable demand or lead time, which requires safety stock and a service-level factor.
What to do instead
Read the stem for the words 'variable', 'uncertain', 'standard deviation', or 'service level'. If present, use reorder point = expected demand during lead time + safety stock (z × σ of demand during lead time). If the stem says demand and lead time are constant, EOQ without safety stock is correct. State the assumption explicitly before computing.
Confusing transportation mode selection criteria with carrier selection criteria, and answering mode-choice questions with rate or service-history arguments that belong to carrier choice.
Both involve picking 'who/how to ship', and candidates conflate the two decisions. Mode choice turns on speed, cost per unit, volume, distance, and product characteristics; carrier choice turns on reliability, coverage, rates, and service record within a mode.
What to do instead
Separate the decisions: first choose the mode based on product and service requirements (air for urgency/high value, rail for bulk/long haul, truck for flexibility, intermodal for balance), then choose a carrier within that mode. If an option argues from a specific carrier's performance while the stem asks about mode, eliminate it.
Treating warehouse layout and materials-handling questions as pure space-utilization problems, ignoring flow, slotting, and pick-path efficiency, so the answer maximizes cube utilization at the expense of throughput.
Utilization is the most intuitive metric and candidates default to 'fill the space'. But high-density storage can lengthen pick paths and slow order fulfillment, which is usually what the exam is testing.
What to do instead
For each layout or handling option, evaluate three things: product flow (receiving to shipping with minimal backtracking), slotting (high-velocity items near shipping and at accessible heights), and handling equipment fit (unit load, aisle width, throughput). Reject options that raise storage density while degrading pick rate unless the stem explicitly prioritizes storage capacity.
Answering order-management questions by optimizing a single order's speed or cost, without considering the order-to-cash cycle, order status visibility, and the effect on customer-perceived reliability.
Candidates focus on fulfillment mechanics and forget that order management is judged on promise accuracy and visibility; a fast shipment that misses the promised date damages the relationship more than a slower accurate one.
What to do instead
For each option, check the order-to-cash path: order capture, credit/availability check, promise date, fulfillment, shipment, invoicing. Ask whether the option improves promise accuracy and status visibility, not just speed. Options that shorten one step while breaking the promise date or hiding status are distractors.
Treating risk and sustainability as separate compliance topics rather than as design variables in network decisions, so questions about resilient or sustainable supply chain design get answered with a single mitigation tactic or a single green initiative.
Risk and sustainability are often studied in isolated chapters, and candidates memorize lists (dual sourcing, buffer stock, carbon reporting) without connecting them to network structure, supplier selection, and cost trade-offs.
What to do instead
For design questions, evaluate options on four axes: probability/impact of disruption, time to recover, cost of mitigation, and environmental/social footprint. Prefer options that address structural exposure (multi-sourcing, geographic diversification, visibility, supplier development) over single-point tactics, and check whether the option names the trade-off it accepts.