Lesson 1Creating derived fields: rolling demand (7/30/90-day), lead time deviation (actual - standard), shipments per unit, cost per unit shippedDis section cover creating derived metrics dat enhance supply chain analysis. Yuh wi compute rolling demand windows, lead time deviations, shipment intensity, an cost per unit shipped fi monitoring an modeling, seen.
Rolling 7/30/90-day demand calculationsLead time deviation an variability metricsShipments per order line an per unitCost per unit shipped an per laneStoring derived fields fi reuse an auditsLesson 2Normalizing categorical fields: supplier_id, transport_mode, warehouse_id; mapping synonyms an encoding fi analysisDis section explain how fi normalize categorical fields like supplier, transport mode, an warehouse identifiers. Yuh wi map synonyms, standardize codes, an encode categories fi modeling an reporting, yuh know.
Standardizing supplier an warehouse IDsCleaning transport_mode an route labelsBuilding synonym an alias mapping tablesHandling slowly changing categorical valuesEncoding categories fi ML an BI toolsLesson 3Reading large CSVs reliably in Excel, Python (pandas) an R (data.table/readr): parsing dates an typesDis section show how fi reliably read large CSVs in Excel, Python, an R widout corrupting types or dates. Yuh wi handle delimiters, encodings, memory limits, chunked loading, an schema definitions fi repeatable ingestion, mi fren.
Configuring delimiters, headers, an encodingsParsing dates, times, an time zones correctlyControlling column types in pandas an RChunked an incremental CSV loadingDealing wid Excel row limits an crashesLesson 4Outlier detection fi time series an cross-sectional fields: z-score, IQR, rolling median filters, an domain thresholdsDis section teach methods fi detect an handle outliers in time series an cross-sectional supply chain data. Yuh wi apply z-score, IQR, rolling statistics, an domain thresholds, den choose fi cap, correct, or exclude values, yuh zeet.
Visual screening of outliers in time seriesZ-score an modified z-score approachesIQR fences an robust spread measuresRolling median an rolling MAD filtersDomain-based thresholds an capping rulesLesson 5Data consistency checks: duplicate rows, negative quantities, mismatched units/currencies, date continuity per SKU-warehouseDis section focus pon validating transactional consistency in supply chain CSVs. Yuh wi detect duplicate records, negative or impossible quantities, unit an currency mismatches, an gaps or overlaps in SKU–warehouse time series, seen.
Detecting an resolving duplicate rowsFlagging negative or impossible quantitiesValidating units of measure an conversionsChecking currency codes an FX alignmentEnsuring date continuity per SKU–warehouseLesson 6Time zone an business calendar adjustments: handling holidays, cutoffs, an business days fi lead time calculationsDis section cover aligning timestamps wid business calendars so lead times, service levels, an cutoffs computed pon comparable business days. Yuh wi adjust fi weekends, regional holidays, an warehouse-specific operating schedules, mi man.
Standardizing time zones across systemsBuilding business day an holiday calendarsModeling shipping an order cutoff timesConverting calendar days to business daysLead time calculation examples in PythonLesson 7Column semantics & metadata mapping: interpreting date, SKU, warehouse, supplier, demand, forecast, inventory, shipments, lead times, costs, flagsDis section focus pon defining column semantics an metadata fi supply chain CSVs. Yuh wi map fields to business concepts, document units an grain, an create dictionaries dat support governance an reuse, yuh hear.
Identifying grain: SKU, location, an timeDefining business meaning of key columnsDocumenting units, currencies, an calendarsCreating an maintaining data dictionariesTagging quality flags an status indicatorsLesson 8Automated data profiling: distributions, missingness matrix, unique counts, value ranges, cardinalityDis section introduce automated profiling of supply chain CSVs. Yuh wi compute distributions, missingness matrices, unique counts, ranges, an cardinality to quickly assess data quality an prioritize cleaning work, seen.
Generating summary statistics at scaleVisualizing missingness matrices an heatmapsAnalyzing value ranges an out-of-bounds dataCardinality checks fi keys an categoriesAutomated profiling wid pandas an R toolsLesson 9Detecting an handling missing values: imputation strategies per column (demand, forecast, inventory, costs) an when to drop rowsDis section explain how fi analyze an treat missing values in demand, forecast, inventory, an cost fields. Yuh wi compare imputation options, design column-specific rules, an decide when dropping rows or segments safer, mi bredda.
Profiling missingness patterns an mechanismsSimple an advanced numeric imputationsImputing categorical an flag variablesColumn-specific rules fi supply chain fieldsCriteria fi dropping rows or time segments
