4-3. Explanation of Data Analysis & System Construction
Logistics System Design Starting from Shipping Data & EIQ / PCB Analysis Simulation
1. Logistics System Characteristics and Basics of Data Analysis
đ ăDiagram ExplanationăDifferences in Characteristics Between Production Logistics and Sales Logistics
The figure above shows the basic structure illustrating the boundary between "Procurement/Production (Production Logistics)" and "Sales (Sales Logistics)" in the overall supply chain.
- Production Logistics (Material Flow): Logistics centered on factories and raw material warehouses. Cargo moves systematically based on production plans, making it easy to control with relatively calm demand fluctuations.
- Sales Logistics (Physical Distribution): Logistics heading from distribution centers to wholesalers, retailers, and end users. Because it is directly influenced by actual "order information," it intrinsically features high SKU diversity, small-lot deliveries, and sudden shipping surges.
In the system construction of distribution centers, advanced data analysis in advance is essential to handle these "variable-quantity, irregular-timing requests unique to sales logistics."
The difference in characteristics between production logistics and sales logistics is extremely large.
What is a Logistics System?
â Complicated Low Technology System
Simply deploying advanced high-tech equipment is not the sole correct answer; optimization of low-tech elements where field operations and IT are intricately intertwined is required.
⥠System Under Numerous Constraints
There are many correct answers in logistics optimization. Furthermore, depending on changes in shipping volumes and cargo packaging conditions, a "correct answer" in one scene may flip to an "incorrect answer" in another, and vice versa.
A distribution center and logistics system is a system with numerous constraints, and there are many correct answers.
However, all designs must be established on top of a "basic system."
The basic system referred to here is a system based on EIQ data and PCB analysis. Because it is based on objective data, whoever logically plans it will reach an optimal, unwavering system. Haphazard system construction not based on data is a major cause of failure.
2. System Approach (Philosophical Thinking)
Systems Thinking
Rather than merely introducing the latest equipment, the "philosophy and mindset" of how to approach system construction is most important. Strategic thinking is required regarding how to decipher collected data and handle rapid economic trends and social environmental changes.
đĄ True Mission of a Distribution Center:
The ultimate mission of a distribution center comes down to "delivering ordered goods to customers accurately by the designated date." Excessive automation, mechanization, or cost reduction itself are merely means, not the purpose or mission itself.
Patterns from Receiving to Shipping and the Importance of "PCB Analysis"
đ ăDiagram ExplanationăTransformation Patterns in PCB Analysis (P=Pallet, C=Case, B=Broken/Piece)
The diagram above illustrates the changes in cargo packaging forms (transformations) from warehouse inbound receiving to outbound shipping.
- P (Pallet): Large-lot packaging unit in pallets.
- C (Case): Medium-lot packaging unit in cardboard cases.
- B (Broken/Bara/Piece): Small-lot packaging unit in pieces or broken cases.
The method of quantitatively capturing the combination patterns of receiving forms (P/C/B) and shipping forms (P/C/B) is called PCB analysis. For example, a site with a high ratio of "received in P, shipped in B (PâB)" requires vast unboxing/piece-picking areas and a high-function WMS. Conversely, a site with a high ratio of "received in C, shipped in C (CâC)" benefits from a logistics system centered on cross-docking and sorters.
Selection of Picking Methods:
ă»Single Picking (Discrete Order Picking): A method where one worker collects ordered items order by order.
ă»Batch Picking (Cluster Picking): A method where items for multiple orders are picked at once and then sorted by shipping destination.
* When adopting Batch Picking, a secondary sorting design (sorters or assortment racks) after picking becomes necessary.
3. Cargo Volume and Suitable Logistics Equipment (Selecting Suitable Material Handling Equipment)
(Original diagram composed by Dr. Shin Suzuki)
đ ăDetailed Diagram ExplanationăSelection Map for Suitable Equipment According to IQ Curve and PCB (Packaging)
This diagram is a systematic chart derived from data analysis (IQ analysis and PCB analysis) to guide the optimal logistics system and storage equipment based on "Shipping Quantity (Q)", "Item Count (I)", and "Packaging Form (P/C/B)".
- A-Rank (High Turnover / Mass Shipping Zone):
Areas where shipping is concentrated. Automated/mechanized equipment such as automated pallet warehouses, mini-load case AS/RS, A-type automatic picking machines, and high-speed sorters are intensively deployed to achieve high-throughput processing. - B-Rank (Medium Turnover Zone):
Equipment balancing storage efficiency and picking workabilityâsuch as reach forklifts with pallet racks, bucket shelves, and carouselsâis selected. - C-Rank (Low Turnover / Multi-SKU Small-Lot Zone):
The number of SKUs is vast, but the shipping volume per item is minimal. Expensive automated equipment is avoided; instead, low-tech operations using fixed medium-duty racks and hand-push carts are adopted to prevent over-investment and optimize overall ROI.
4. How to Read Data and Data Positioning Techniques
When conducting data analysis, the most common trap to fall into is "treating a single day's gathered data directly as the company-wide standard value."
đ ăDiagram Explanationă4 Key Mindsets for Successful Data Analysis
- Observe Data Rather Than Just Looking: Do not merely look at numbers; observe and decipher the underlying operational movements in the field. By accumulating case studies from other companies, you can gain deep insights into the meaning of figures.
- Do Not Hold Preconceived Notions: Eliminate equipment-driven thoughts such as "I want to use that machine" or "Let's copy another company." A logistics system should be built uniquely based on EIQ data.
- Do Not Collect Too Much Data / A 10% Difference is Margin of Error: Avoid getting overly bogged down in extreme precision. Narrow down only the necessary data and swiftly enter evaluation. Treat a ~10% fluctuation in logistics data as a margin of error, prioritizing quickly grasping the overall trend.
- Read Data by Packaging Form (PCB): A figure of "100 cases" translates to "4 pallets + 4 cases" when evaluated under 1 pallet = 24 cases. Possessing a PCB analysis perspective reveals specific workload requirements for actual pallet handling and case picking.
Relative Positioning of Surveyed Days (Understanding Maximum and Average)
Accurate system construction is impossible without understanding: "Where does the cargo volume of a specific single day analyzed stand relative to the annual maximum and average values?"
đ ăDiagram ExplanationăFluctuation Comparison Analysis: Analysis Day (Feb 23) vs. Peak Day (May 9)
The graph above illustrates daily volume fluctuations during the operational period (Apr 25âMay 9) compared to a specific analysis day (Feb 23).
- Analysis Day (Feb 23) Data: Shipping volume of 36,402 balls (of which frozen food accounts for 24,466 balls), operating time 7.7 hours. Shown as a red reference line (24,466 balls) on the graph.
- Peak Day (May 9) Data: Frozen food shipping volume reached 27,238 balls, jumping expected operating time to 8.2 hours.
Practical Insight: If system capacity and regular staffing are designed solely around the baseline volume (24,466 balls) of the analysis day (Feb 23), the logistics system will break down on peak days like May 9 (27,238 balls). When evaluating analysis data, it is essential to capture where the baseline day stands against annual peaks, building in equipment buffers and shift planning (overtime / temporary labor) capable of covering peak days (8.2 hours operating time).
Analysis of Data File Field Definitions and Operational Rules
đ ăDiagram ExplanationăData Field Definitions and Picking/Consolidation Logic
The figure above shows an example structure of shipping transaction data extracted from a WMS (Warehouse Management System) along with definitions of operational rules.
- Main Fields: Shipping Date, Shipper Department, Agency (Orderer), Destination (End Customer), Order Number, Shipping Number, Inventory Item Code/Name, Lot Number, Serial Number, Shipping Quantity, etc., are defined.
- Picking & Consolidation Rules:
â "Pick Inventory Item Code per End Customer Number, scan the Lot Number and Serial Number of the picked items, and append them to shipping information."
⥠"Items picked per End Customer Number are consolidated per shipping destination before dispatch."
Practical Insight: In system construction, you must decipher not just numeric data like "what item and how many items to send," but also the operational logic behind data structuresâsuch as "at what unit (End Customer) to pick, and at what stage to consolidate orders." Clarifying SKUs and consolidation units allows precise design of required packing stations and barcode scanner operational flows.
5. Overview and Purpose of EIQ Analysis
đ ăDiagram Explanationă7 Approaches of EIQ Research and Roles of Practitioners vs. Researchers
The chart above outlines 7 research themes related to the EIQ method and defines whether each should be led by "Practitioners" or "Researchers."
- Themes Led by Practitioners (On-site Operations & System Construction):
1. Research on simplified status analysis using EIQ methods.
2. Research on selection methods for operations, information, and equipment matching EIQ characteristics.
6. Research for designing logistics systems from EIQ analysis and functional requirements.
7. Research on DATA processing methods to improve outbound efficiency. - Themes Led by Researchers (Theoretical Models & Data Conversion):
3. Research on methods and expressions to derive EIQ characteristics from DATA sources.
4. Research on generating DATA sources from EIQ characteristic patterns.
5. Research on estimating outbound method loads through DATA source analysis.
Practical Insight: Practitioners in charge of on-site system construction should not merely chase aggregation logic (researcher domain), but focus on "what material handling equipment and operations to select and design based on shipping characteristics derived from EIQ data."
EIQ Analysis Program and Basic Steps
đ ăDetailed Diagram ExplanationăIntegrated Flow from Data Analysis to System Design and Logistics Cost Calculation
The diagram above illustrates a system mechanism that starts from EIQ analysis data to seamlessly simulate distribution center spatial/handling design up to final investment cost calculation.
- EIQ Analysis & EIQ Applications: Predicts theoretical inventory, reorder points, and inbound volumes from EIQ data, automatically generating missing data rescue, EIQ matrices, and curves.
- System Design Data Section (Shipping Analysis & Handling Design): Levels hourly shipping volume, separates Case/Broken/Pallet units (PCB Analysis), calculates handling loads for inbound, outbound, replenishment, and piece picking, and sets inspection/packing counts and berth requirements.
- Physical Area Design (Inbound, Storage, Shipping Areas): Calculates required areas for receiving, storage, and shipping from required pallet counts, estimates staffing, and auto-generates equipment layout drawings.
- Logistics Cost Calculation Section (Sales Tool): Aggregates costs for buildings, equipment, materials, labor, and delivery, simulating Return on Investment (ROI), lease fees, and depreciation.
Practical Insight: Excellent data analysis does not end with creating reports; it demonstrates true value when utilized as the foundation for system construction that automatically links physical required areas (storage/shipping), required labor, material handling machinery, and ROI/cost estimates as shown in the flow.
đ ăDiagram Explanationă6 Challenges Solved by EIQ Analysis
EIQ analysis provides quantitative answers to the following questions:
- What storage equipment (pallet racks, case shelves, automated warehouses) is required?
- Which picking method is optimalâsingle discrete order picking or batch cluster picking?
- Which layout flow lines minimize walking distances most effectively?
- What material handling equipment (conveyors, sorters, AGVs) should be introduced?
- What is the optimal staffing layout matching hourly shipping fluctuations?
- How scalable is the system against future volume growth?
6. How to Read EIQ Indicators and Required Data
1. Item Count Ratio (I Analysis): The proportion of SKUs that are actively moving (shipped) out of total registered SKUs.
2. Volume Ratio (Q Analysis): Concentration of shipping quantities (Pareto principle: whether top 20% items account for 80% of total volume).
3. Destination Count Ratio (E Analysis): Order frequency and lot size variance across delivery destinations (customers).
4. Shipping Volume Ratio (EQ Analysis): Shipping volume per delivery destination (large bulk vs. small-lot high-frequency).
Required Survey Items for EIQ Analysis and PCB Analysis
7. EIQ Matrix and Application to Layout Design
Translating Analysis Results into Layout Drawings (Spatial Design)
đ ăDiagram ExplanationăReflecting Data Analysis (EIQ & PCB) into Layouts
The ultimate goal of data analysis is not to produce an analytical report, but to translate findings into an actual "distribution center layout drawing (spatial arrangement)."
- High Turnover Zone (A-Rank Ă Case/Pallet Form): Positioned immediately near shipping docks, directly connected to forklifts or automated sorters.
- Medium Turnover Zone (B-Rank Ă Piece Form): Medium-duty shelves or Digital Picking Systems (DPS) are deployed to optimize walking flow lines.
- Low Turnover Zone (C-Rank Ă Long-term Storage): Positioned deep in the warehouse or on high-level mezzanines to maximize floor space efficiency (volumetric utilization).
By following a series of analytical stepsâData Analysis, PCB Analysis, and EIQ Analysisâthe construction of a rational, highly functional, and ideal logistics system is successfully achieved.