CSOWM cheat sheet

Total landed cost = unit price + freight + duties/tariffs + insurance + inventory carrying cost + handling + risk cost; compare sourcing options only on this sum, not on unit price alone.

Center-of-gravity (weighted centroid) for a single facility: X = Σ(volume_i × x_i) / Σvolume_i, Y = Σ(volume_i × y_i) / Σvolume_i; use as a starting point, then adjust for road network, labor, and land.

Number of warehouses trade-off: adding a node reduces outbound distance and lead time but raises safety stock, inbound freight, and facility fixed cost; optimal count is where marginal service benefit equals marginal cost.

Postponement (delayed differentiation) shifts product differentiation downstream, cutting finished-goods SKU inventory at the cost of higher per-unit assembly/configuration expense.

Push vs pull decoupling point: the push-pull boundary sits where forecast-driven production meets order-driven fulfillment; longer forecast-driven segments mean more forecast risk.

SCOR framework processes: Plan, Source, Make, Deliver, Return, Enable; use it to map scope before designing metrics.

Perfect order rate = (orders delivered complete × on time × damage-free × correct documentation) / total orders; it is multiplicative, so each component failure zeroes the order.

Network design decisions are capital-intensive and long-lived (5-10+ years); distribution and transport decisions are tactical (quarterly-annual) and can be revised faster.

Moving average of n periods = sum of last n actuals / n; higher n gives smoother output and slower response to level shifts.

Simple exponential smoothing: F(t+1) = α·A(t) + (1 − α)·F(t); higher α weights recent actuals more heavily.

Holt's linear trend method adds a trend term: level L(t) = α·A(t) + (1 − α)(L(t−1) + T(t−1)); trend T(t) = β(L(t) − L(t−1)) + (1 − β)T(t−1).

Winter's triple exponential smoothing adds a seasonal index (additive or multiplicative) with smoothing constant γ; use when seasonality repeats with stable amplitude.

Mean absolute deviation (MAD) = Σ|A − F| / n; mean absolute percentage error (MAPE) = (Σ|A − F| / A) / n × 100, undefined when actuals are zero.

Bias (mean forecast error) = Σ(A − F) / n; near-zero bias means no systematic over- or under-forecast, but does not imply low MAPE.

Tracking signal = running sum of forecast errors / MAD; a value outside roughly ±4 signals a forecast model that needs re-fitting.

Forecast hierarchy reconciliation: aggregate SKU-level forecasts to family level and compare with the family forecast; reconcile bottom-up or top-down before committing to supply plans.

Coefficient of variation (CV) = standard deviation / mean; high-CV items are candidates for make-to-order or postponement rather than forecast-driven stock.

Kraljic matrix axes: profit impact (spend, value contribution) and supply risk; quadrants are leverage, strategic, bottleneck, and non-critical, each with a distinct sourcing approach.

Supplier scorecard weights should sum to 100%: typical dimensions are quality (PPM defects), delivery (on-time-in-full), cost, responsiveness, and sustainability/compliance.

Total cost of ownership includes price, freight, duties, quality inspection, rework, warranty, inventory, and end-of-life disposal; lowest bid rarely equals lowest TCO.

Single vs dual sourcing: single sourcing maximizes volume leverage and simplicity but concentrates risk; dual sourcing trades price for continuity and competitive tension.

Early supplier involvement (ESI) moves design collaboration upstream; the earlier a supplier influences design, the lower the cost of subsequent engineering changes.

Contract forms: fixed-price transfers cost risk to the supplier; cost-plus transfers risk to the buyer; incentive contracts share risk against a target cost and share ratio.

Supplier segmentation by spend (ABC/Pareto): a small share of suppliers typically accounts for the majority of spend and warrants formal contracts and reviews.

Purchase price variance (PPV) = (actual price − standard price) × quantity purchased; it measures price performance only and ignores quality and delivery cost.

Economic order quantity: EOQ = √(2DS / H), where D = annual demand, S = order cost per order, H = annual holding cost per unit.

Reorder point (continuous review) = demand during lead time + safety stock; for variable demand and lead time, use the combined standard deviation of demand over lead time.

Safety stock = z × σ(demand over lead time); z is the standard normal quantile for the target cycle service level (z ≈ 1.28 for 90%, 1.65 for 95%, 2.33 for 99%).

Periodic review order-up-to level = demand over (review period + lead time) + safety stock; longer review intervals require larger safety stock.

Inventory turnover = cost of goods sold / average inventory value; days of supply = 365 / turnover.

ABC classification by annual usage value (unit cost × annual volume): A items justify tight cycle counting and frequent review; C items justify simple rules.

Fill rate and cycle service level differ: cycle service level is the probability of no stockout per replenishment cycle; fill rate is the fraction of demand met from stock.

Carrying cost is usually 15-30% of unit value per year and includes capital, storage, insurance, obsolescence, and shrinkage.

Bullwhip effect is amplified order variability moving upstream; causes include demand signal distortion, order batching, price promotions, and shortage gaming.

Cube utilization = occupied storage volume / total available storage volume; floor utilization = occupied floor area / total floor area.

Slotting rule: assign highest pick-frequency SKUs to the golden zone (roughly waist-to-shoulder height) and nearest to the dispatch/forward area to minimize travel.

ABC velocity slotting: A items in the most accessible, shortest-travel locations; C items in high, low, or remote locations.

Travel time typically dominates order-picking labor (often around half of picking time), so layout changes that shorten travel yield the largest productivity gains.

Forward (pick-face) area sized to a few days of demand; reserve area holds bulk. Too-large forward areas waste space; too-small ones cause frequent replenishment.

Aisle width trade-off: narrow aisles raise storage density but require specialized equipment and slow travel; wide aisles suit high-throughput, multi-shift operations.

Cross-docking layout requires staged inbound and outbound doors with minimal put-away; flow-through design avoids storage entirely for eligible freight.

Slotting should be re-run after assortment or demand changes; seasonal re-slotting prevents drift from the original velocity profile.

Receiving flow: appointment/scheduling → unload → count and inspect → reconcile against PO/ASN → put-away; discrepancies should be recorded before put-away.

Put-away strategies: directed put-away (system-assigned location) maximizes control; random put-away maximizes space utilization but requires reliable location tracking.

Order picking methods: discrete (one order at a time), batch (multiple orders picked together), zone (picker confined to an area), and wave picking (released in scheduled waves).

Pick-to-light and voice-directed picking reduce errors and training time; RF scanning enforces confirmation at each scan point.

Labor standards: units per labor hour (UPLH) and lines per hour are common productivity measures; compare against engineered standards, not across dissimilar facilities.

Cycle counting counts a subset of locations on a rotating schedule; A items counted more often; variance above tolerance triggers root-cause investigation.

Loading dock scheduling smooths inbound and outbound arrivals; dock door utilization = dock-occupied time / available dock time.

Materials handling equipment selection: counterbalance forklifts for general loads, reach trucks for narrow aisles and high racks, order pickers for piece picking at height.

Freight class and density: density = weight / volume; higher density generally yields lower cost per unit weight because trailers cube out less often.

Truckload (TL) vs less-than-truckload (LTL): TL uses a full trailer for one shipper; LTL consolidates multiple shippers' freight and adds terminal handling and transit time.

Cost per mile = total trip cost / miles; cost per hundredweight (cwt) = freight cost / (weight in lb / 100) allows comparison across shipment sizes.

Mode selection trade-off: air fastest and most expensive per ton-mile; ocean and rail cheapest per ton-mile but slowest and with larger minimum quantities.

Incoterms allocate cost, risk, and customs responsibility: FOB risk passes at shipment on board; CIF seller pays cost, insurance, freight to destination but risk passes at shipment; DDP seller bears duty and delivery.

Freight bill audit checks charges against the contracted tariff, accessorials, and weight/dimension; fuel surcharges are typically indexed to a published diesel price.

Route optimization objective is usually minimizing total distance or cost subject to time windows, vehicle capacity, and driver hours rules.

Cube out vs weigh out: a trailer that fills volume before reaching weight limit is cube-limited; loading plans should maximize both.

WMS manages execution inside the four walls (receiving, put-away, picking, shipping); TMS plans and executes freight movement; ERP holds financial and master data.

EDI standards: X12 in North America, EDIFACT internationally; common transaction sets include 850 (purchase order), 856 (ASN), 810 (invoice), 214 (shipment status).

Barcode symbologies: Code 128 for dense alphanumeric data; GS1-128 encodes application identifiers such as SSCC and lot; QR and Data Matrix are 2D codes.

RFID enables non-line-of-sight, bulk reads without individual scanning; passive tags have no battery and derive power from the reader field.

Data quality dimensions for logistics analytics: completeness, accuracy, timeliness, consistency, and uniqueness; master data errors propagate into every downstream plan.

KPI dashboard design: pair a lagging outcome (perfect order rate, OTIF) with leading drivers (dock-to-stock time, pick accuracy) to make the metric actionable.

Control tower visibility combines carrier, supplier, and internal event data into exception alerts; value comes from exception management, not raw tracking volume.

Forecast/planning systems separate master data, transaction data, and analytical models; integration latency between WMS and ERP affects inventory accuracy timing.

Risk assessment: likelihood × impact scores prioritize mitigations; residual risk is what remains after controls are applied.

Business continuity planning requires a documented RTO (recovery time objective) and RPO (recovery point objective) for critical logistics systems and nodes.

Single-source and single-node dependencies are concentration risks; mitigation options include alternate suppliers, buffer stock, and dual-route transport.

OSHA-style safety hierarchy of controls: elimination, substitution, engineering controls, administrative controls, then PPE as the last line of defense.

Forklift safety fundamentals: pre-shift inspection, pedestrian separation, load capacity plate limits, and load center awareness when carrying off-center loads.

Hazmat shipping requires classification, UN number, proper shipping name, packing group, labels/placards, and a shipping paper; training is required for handlers.

Sustainability metrics: CO2e per shipment or per ton-mile; scope 1 (direct fuel), scope 2 (purchased energy), scope 3 (supply chain and transport) emissions.

Customs compliance: correct HS classification, declared value, country of origin, and record retention support duty calculation and preferential trade program eligibility.

Regulatory recordkeeping (e.g., hours-of-service, training, hazmat shipping papers) must be retained for defined periods and available for inspection.

Total landed cost = unit price + freight + duties/tariffs + insurance + inventory carrying cost + handling + risk cost; compare sourcing options only on this sum, not on unit price alone.

Center-of-gravity (weighted centroid) for a single facility: X = Σ(volume_i × x_i) / Σvolume_i, Y = Σ(volume_i × y_i) / Σvolume_i; use as a starting point, then adjust for road network, labor, and land.

Number of warehouses trade-off: adding a node reduces outbound distance and lead time but raises safety stock, inbound freight, and facility fixed cost; optimal count is where marginal service benefit equals marginal cost.

Postponement (delayed differentiation) shifts product differentiation downstream, cutting finished-goods SKU inventory at the cost of higher per-unit assembly/configuration expense.

Push vs pull decoupling point: the push-pull boundary sits where forecast-driven production meets order-driven fulfillment; longer forecast-driven segments mean more forecast risk.

SCOR framework processes: Plan, Source, Make, Deliver, Return, Enable; use it to map scope before designing metrics.

Perfect order rate = (orders delivered complete × on time × damage-free × correct documentation) / total orders; it is multiplicative, so each component failure zeroes the order.

Network design decisions are capital-intensive and long-lived (5-10+ years); distribution and transport decisions are tactical (quarterly-annual) and can be revised faster.

Moving average of n periods = sum of last n actuals / n; higher n gives smoother output and slower response to level shifts.

Simple exponential smoothing: F(t+1) = α·A(t) + (1 − α)·F(t); higher α weights recent actuals more heavily.

Holt's linear trend method adds a trend term: level L(t) = α·A(t) + (1 − α)(L(t−1) + T(t−1)); trend T(t) = β(L(t) − L(t−1)) + (1 − β)T(t−1).

Winter's triple exponential smoothing adds a seasonal index (additive or multiplicative) with smoothing constant γ; use when seasonality repeats with stable amplitude.

Mean absolute deviation (MAD) = Σ|A − F| / n; mean absolute percentage error (MAPE) = (Σ|A − F| / A) / n × 100, undefined when actuals are zero.

Bias (mean forecast error) = Σ(A − F) / n; near-zero bias means no systematic over- or under-forecast, but does not imply low MAPE.

Tracking signal = running sum of forecast errors / MAD; a value outside roughly ±4 signals a forecast model that needs re-fitting.

Forecast hierarchy reconciliation: aggregate SKU-level forecasts to family level and compare with the family forecast; reconcile bottom-up or top-down before committing to supply plans.

Coefficient of variation (CV) = standard deviation / mean; high-CV items are candidates for make-to-order or postponement rather than forecast-driven stock.

Kraljic matrix axes: profit impact (spend, value contribution) and supply risk; quadrants are leverage, strategic, bottleneck, and non-critical, each with a distinct sourcing approach.

Supplier scorecard weights should sum to 100%: typical dimensions are quality (PPM defects), delivery (on-time-in-full), cost, responsiveness, and sustainability/compliance.

Total cost of ownership includes price, freight, duties, quality inspection, rework, warranty, inventory, and end-of-life disposal; lowest bid rarely equals lowest TCO.

Single vs dual sourcing: single sourcing maximizes volume leverage and simplicity but concentrates risk; dual sourcing trades price for continuity and competitive tension.

Early supplier involvement (ESI) moves design collaboration upstream; the earlier a supplier influences design, the lower the cost of subsequent engineering changes.

Contract forms: fixed-price transfers cost risk to the supplier; cost-plus transfers risk to the buyer; incentive contracts share risk against a target cost and share ratio.

Supplier segmentation by spend (ABC/Pareto): a small share of suppliers typically accounts for the majority of spend and warrants formal contracts and reviews.

Purchase price variance (PPV) = (actual price − standard price) × quantity purchased; it measures price performance only and ignores quality and delivery cost.

Economic order quantity: EOQ = √(2DS / H), where D = annual demand, S = order cost per order, H = annual holding cost per unit.

Reorder point (continuous review) = demand during lead time + safety stock; for variable demand and lead time, use the combined standard deviation of demand over lead time.

Safety stock = z × σ(demand over lead time); z is the standard normal quantile for the target cycle service level (z ≈ 1.28 for 90%, 1.65 for 95%, 2.33 for 99%).

Periodic review order-up-to level = demand over (review period + lead time) + safety stock; longer review intervals require larger safety stock.

Inventory turnover = cost of goods sold / average inventory value; days of supply = 365 / turnover.

ABC classification by annual usage value (unit cost × annual volume): A items justify tight cycle counting and frequent review; C items justify simple rules.

Fill rate and cycle service level differ: cycle service level is the probability of no stockout per replenishment cycle; fill rate is the fraction of demand met from stock.

Carrying cost is usually 15-30% of unit value per year and includes capital, storage, insurance, obsolescence, and shrinkage.

Bullwhip effect is amplified order variability moving upstream; causes include demand signal distortion, order batching, price promotions, and shortage gaming.

Cube utilization = occupied storage volume / total available storage volume; floor utilization = occupied floor area / total floor area.

Slotting rule: assign highest pick-frequency SKUs to the golden zone (roughly waist-to-shoulder height) and nearest to the dispatch/forward area to minimize travel.

ABC velocity slotting: A items in the most accessible, shortest-travel locations; C items in high, low, or remote locations.

Travel time typically dominates order-picking labor (often around half of picking time), so layout changes that shorten travel yield the largest productivity gains.

Forward (pick-face) area sized to a few days of demand; reserve area holds bulk. Too-large forward areas waste space; too-small ones cause frequent replenishment.

Aisle width trade-off: narrow aisles raise storage density but require specialized equipment and slow travel; wide aisles suit high-throughput, multi-shift operations.

Cross-docking layout requires staged inbound and outbound doors with minimal put-away; flow-through design avoids storage entirely for eligible freight.

Slotting should be re-run after assortment or demand changes; seasonal re-slotting prevents drift from the original velocity profile.

Receiving flow: appointment/scheduling → unload → count and inspect → reconcile against PO/ASN → put-away; discrepancies should be recorded before put-away.

Put-away strategies: directed put-away (system-assigned location) maximizes control; random put-away maximizes space utilization but requires reliable location tracking.

Order picking methods: discrete (one order at a time), batch (multiple orders picked together), zone (picker confined to an area), and wave picking (released in scheduled waves).

Pick-to-light and voice-directed picking reduce errors and training time; RF scanning enforces confirmation at each scan point.

Labor standards: units per labor hour (UPLH) and lines per hour are common productivity measures; compare against engineered standards, not across dissimilar facilities.

Cycle counting counts a subset of locations on a rotating schedule; A items counted more often; variance above tolerance triggers root-cause investigation.

Loading dock scheduling smooths inbound and outbound arrivals; dock door utilization = dock-occupied time / available dock time.

Materials handling equipment selection: counterbalance forklifts for general loads, reach trucks for narrow aisles and high racks, order pickers for piece picking at height.

Freight class and density: density = weight / volume; higher density generally yields lower cost per unit weight because trailers cube out less often.

Truckload (TL) vs less-than-truckload (LTL): TL uses a full trailer for one shipper; LTL consolidates multiple shippers' freight and adds terminal handling and transit time.

Cost per mile = total trip cost / miles; cost per hundredweight (cwt) = freight cost / (weight in lb / 100) allows comparison across shipment sizes.

Mode selection trade-off: air fastest and most expensive per ton-mile; ocean and rail cheapest per ton-mile but slowest and with larger minimum quantities.

Incoterms allocate cost, risk, and customs responsibility: FOB risk passes at shipment on board; CIF seller pays cost, insurance, freight to destination but risk passes at shipment; DDP seller bears duty and delivery.

Freight bill audit checks charges against the contracted tariff, accessorials, and weight/dimension; fuel surcharges are typically indexed to a published diesel price.

Route optimization objective is usually minimizing total distance or cost subject to time windows, vehicle capacity, and driver hours rules.

Cube out vs weigh out: a trailer that fills volume before reaching weight limit is cube-limited; loading plans should maximize both.

WMS manages execution inside the four walls (receiving, put-away, picking, shipping); TMS plans and executes freight movement; ERP holds financial and master data.

EDI standards: X12 in North America, EDIFACT internationally; common transaction sets include 850 (purchase order), 856 (ASN), 810 (invoice), 214 (shipment status).

Barcode symbologies: Code 128 for dense alphanumeric data; GS1-128 encodes application identifiers such as SSCC and lot; QR and Data Matrix are 2D codes.

RFID enables non-line-of-sight, bulk reads without individual scanning; passive tags have no battery and derive power from the reader field.

Data quality dimensions for logistics analytics: completeness, accuracy, timeliness, consistency, and uniqueness; master data errors propagate into every downstream plan.

KPI dashboard design: pair a lagging outcome (perfect order rate, OTIF) with leading drivers (dock-to-stock time, pick accuracy) to make the metric actionable.

Control tower visibility combines carrier, supplier, and internal event data into exception alerts; value comes from exception management, not raw tracking volume.

Forecast/planning systems separate master data, transaction data, and analytical models; integration latency between WMS and ERP affects inventory accuracy timing.

Risk assessment: likelihood × impact scores prioritize mitigations; residual risk is what remains after controls are applied.

Business continuity planning requires a documented RTO (recovery time objective) and RPO (recovery point objective) for critical logistics systems and nodes.

Single-source and single-node dependencies are concentration risks; mitigation options include alternate suppliers, buffer stock, and dual-route transport.

OSHA-style safety hierarchy of controls: elimination, substitution, engineering controls, administrative controls, then PPE as the last line of defense.

Forklift safety fundamentals: pre-shift inspection, pedestrian separation, load capacity plate limits, and load center awareness when carrying off-center loads.

Hazmat shipping requires classification, UN number, proper shipping name, packing group, labels/placards, and a shipping paper; training is required for handlers.

Sustainability metrics: CO2e per shipment or per ton-mile; scope 1 (direct fuel), scope 2 (purchased energy), scope 3 (supply chain and transport) emissions.

Customs compliance: correct HS classification, declared value, country of origin, and record retention support duty calculation and preferential trade program eligibility.

Regulatory recordkeeping (e.g., hours-of-service, training, hazmat shipping papers) must be retained for defined periods and available for inspection.