2. Data Analysis Methods and Procedures
2-6. Required Items and Logic for DATA Analysis
Purpose of Data Analysis
The primary purpose of data analysis is to accurately express actual logistics operations in numerical facts and to quantitatively demonstrate how equipment, workforce, space, and operating hours will be optimized under proposed system implementations.
1) Elements of Shipment Data (EIQ + Extended Fields)
While EIQ Analysis (E: Destination, I: Item, Q: Quantity) forms the foundation of facility design, simple statements such as "10 units of Item B shipped to Customer A" are insufficient for advanced engineering. Tera Calculation employs an import format consisting of the following core and extended fields:
| Field Name | Description & Operational Role |
|---|---|
| Line Sequence No. | Sequential row index number assigned to each record in shipment data. |
| Shipment Date & Time | Date and time slot of shipment. Used for hourly volume calculations and fluctuation analysis. |
| Destination (Entry) | Name or code of customer, store, or distribution center receiving the goods. |
| Item | Product name or SKU code enabling distinct Stock Keeping Unit identification. |
| Piece Count (Quantity) | Minimum operational and inventory management unit (SKU quantity). |
| Case Quantity (Pack Size) | Master attribute used to convert piece counts into full cases and segregate case/piece orders. |
| Cases per Pallet (Ti-Hi) | Number of cases stacked on one pallet (PL). Used for pallet conversion. |
| Case Volume & Weight | External volume (m³) and gross weight (kg) per case. Used for volume/weight conversions. |
| Item / Shipment Category | Filter flags used to include or exclude specific ranges of data (e.g., direct dropship orders). |
| Shipping Route | Route name or group code managing destinations loaded onto the same delivery truck. |
2) Data Scale and Simple Aggregation
An example of data scale in a standard benchmark model (e.g., a B2B fulfillment center for small electronics):
- Total Record Count: Approx. 60,000 rows (records)
- Scale Classification: Upper-range medium facility to entry-level large facility
- Role of Aggregation: Simple totals are processed initially to evaluate overall metrics, such as total items handled, total destinations, total cases, volume, and gross tonnage.
3) Shipment Volume Fluctuations and Economic Facility Sizing
Shipment volumes fluctuate daily based on seasons, months, day of week, weather, sales, and store openings. This is known as "Shipment Fluctuation."
[Key Principles of Economical Sizing]
- Pitfalls of Peak-Day Sizing: Designing warehouse capacity and automation based on maximum peak-day volumes leads to excessive idle space and unused machinery during normal periods.
- Selecting the Target Day: In Tera Calculation, a standardized representative day (e.g., Thursday rather than peak day) is chosen as the design target.
- Consensus on Peak Handling: Volumes exceeding target capacity during peaks are managed via extended working hours, temporary staffing, third-party warehousing, or advance picking operations. Obtaining cross-departmental agreement (sales, purchasing) is essential.
4) Data Processing Logic (Tera Calculation 0)
After raw shipment data is imported, automated processing executes the following steps to construct the core "T200" Consolidated Table:
- Unit Conversions from Piece Counts:
- Case Equivalent ＝ Piece Count ÷ Case Quantity
- PL Equivalent ＝ Case Equivalent ÷ Cases per Pallet
- Volume Equivalent ＝ Case Equivalent × Case Volume ／ Weight Equivalent ＝ Case Equivalent × Case Weight
- Automatic Segregation of Case and Piece Orders:
- Less than 1 Case (Case Eq. ＜ 1) ──＞ Piece Shipping (Picked from flow racks)
- 1+ Case without Remainder (Case Eq. ≥ 1 & Remainder ＝ 0) ──＞ Case Shipping (Retrieved from stock pallets)
- 1+ Case with Remainder (e.g., order of 9 units with pack size 5) ──＞ Split into 1 Case Shipping and 4 Piece Shipping
- 5-Stage Ranking (Expanded EIQ Analysis):
Expands traditional ABC analysis (3 categories) into 5 tiers: A1 (50%), A2 (20%), B (15%), C (10%), D (5%).- Tera Calculation 1 (Operational Capacity): Uses "Case Equivalent" for case shipping and "Line Count (Frequency)" for piece shipping as ranking keys.
- Tera Calculation 2 (Storage Capacity): Uses "PL Equivalent" for both case and piece shipping to calculate storage space.
- Creation of Consolidated Table "T200": Integrates conversions and ranking attributes into "T200" to enable high-speed SQL queries and analysis.
5) Operational Definitions and Practical Guidelines
- Records vs. Retrieval Frequency: If 1 order line contains "1 case + 3 pieces", it counts as 1 record, but represents 2 retrieval actions (1 case retrieval + 1 piece pick). Tera Calculation accurately counts this as 2 retrieval operations to evaluate labor load.
- Terminology Definitions:
- Storage In / Retrieval: Storing goods into or retrieving goods from bulk storage racks.
- Sorting (Seeding): Allocating bulk-retrieved items into individual order/destination slots.
- Picking (Piece Picking): Picking individual items from pick-face shelving according to orders.
- Upstream Time Shifts: Upstream processes (picking, replenishment) must be scheduled earlier than final departure times, accounting for operational time lags in facility design.