Zhuojin Pipe Fitting Co., LtdDesign and Implementation of Statistical Graphics System TIGS
Literature Overview
This paper published in Computer Engineering (Vol. 18, Issue 6, 1992, pp. 10-15) by Cai Shijie, Yin Jianwen, Ge Ruding, Li Binyu, and Dong Yi from Nanjing University describes the design philosophy, user interface architecture, and multilingual development approach of the Statistical Graphics System (TIGS). Although this work is primarily a computer science contribution rather than a direct welding or materials science study, the system's design principles and statistical visualization capabilities are directly applicable to quality control, process monitoring, and data analysis in steel pipe and pipe fitting manufacturing environments.
Core Technical Contributions
Design Philosophy
TIGS was conceived as a comprehensive statistical graphics system that enables users to generate a wide variety of statistical plots and charts through an intuitive, menu-driven interface. The system's design philosophy emphasizes accessibility for non-programmers while maintaining the flexibility required for complex statistical analysis. The key design principles include:
- Operation-oriented interface: The user interface is structured around common statistical operations rather than individual chart types, reducing the learning curve for quality engineers and production personnel.
- Undo mechanism: The system incorporates a robust undo mechanism that allows users to reverse operations without losing progress, which is critical for exploratory data analysis in quality control settings.
- Menu-driven architecture: All functionality is accessible through hierarchical menus, eliminating the need for command-line input or programming knowledge.
User Interface Architecture
The TIGS interface employs a menu-driven design with operation-oriented and rollback mechanisms:
| Interface Feature | Description | Quality Control Application |
|---|---|---|
| Operation-oriented menus | Grouped by statistical operation type | Quick access to common QC chart types |
| Undo mechanism | Reversible operation history | Safe exploratory analysis |
| Menu hierarchy | Logical grouping of functions | Efficient navigation for complex analyses |
| Multilingual support | Multiple language interface options | International manufacturing environments |
Multilingual Development Approach
The system was developed using multiple programming languages, which enabled both efficient implementation and broad compatibility. This multilingual development approach allowed the system to leverage the strengths of different languages for different components, such as using high-level languages for the user interface and lower-level languages for computational-intensive statistical operations.
Application to Steel Pipe Manufacturing Quality Control
While TIGS was not developed specifically for welding or pipe manufacturing, its statistical graphics capabilities are directly relevant to several quality control activities in the industry:
Statistical Process Control (SPC)
| SPC Application | TIGS Capability | Manufacturing Relevance |
|---|---|---|
| Control charts (X-bar, R, S) | Statistical plot generation | Monitoring weld dimensions, pipe dimensions |
| Histograms | Distribution visualization | Analyzing dimensional tolerance distributions |
| Scatter plots | Correlation analysis | Investigating parameter-quality relationships |
| Pareto charts | Defect prioritization | Identifying dominant defect types in pipe production |
| Box plots | Outlier detection | Identifying anomalous welding parameters |
Process Capability Analysis
The statistical visualization capabilities of TIGS can support process capability studies for critical manufacturing parameters such as:
- Weld bead geometry (width, reinforcement, penetration depth)
- Pipe dimensional tolerances (diameter, wall thickness, roundness)
- Mechanical property distributions (tensile strength, hardness, impact energy)
- NDT result distributions (porosity size, slag inclusion frequency)
Engineering Practice Integration
For a steel pipe manufacturing facility implementing a statistical quality control program, TIGS represents an early example of the type of statistical software that enables data-driven decision making:
- Welding parameter optimization: Scatter plots and regression analysis can identify optimal welding current, voltage, travel speed, and gas flow rate combinations for specific pipe geometries.
- Defect trend analysis: Control charts can detect shifts in defect rates over time, enabling proactive process adjustments before quality escapes.
- Supplier qualification: Statistical analysis of incoming material properties (chemical composition, mechanical properties, dimensional tolerances) supports objective supplier evaluation.
- Equipment performance monitoring: Trend analysis of welding equipment parameters (arc voltage stability, torch oscillation, wire feed consistency) enables predictive maintenance.
Key Questions and Reflections
Several aspects of the TIGS system design raise questions relevant to modern quality control software development:
- How does the operation-oriented interface design compare to modern dashboard-based quality control interfaces in terms of user efficiency?
- What is the appropriate level of statistical sophistication for manufacturing quality engineers—sufficient to perform meaningful analysis without requiring statistical expertise?
- How can the undo mechanism be extended to support version control of quality analysis results, which is important for audit trail purposes in regulated industries?
- What are the computational limitations of the multilingual development approach when applied to large datasets typical of modern manufacturing environments?
Study Insights and Implications
The TIGS system represents an important early effort in making statistical analysis accessible to non-specialists through intuitive interface design. The operation-oriented and menu-driven approach, combined with the undo mechanism, established design patterns that remain relevant in modern quality control software. For steel pipe and pipe fitting manufacturers, the availability of accessible statistical visualization tools is essential for implementing effective quality control programs based on statistical process control principles. The multilingual development approach also reflects the growing internationalization of manufacturing, where quality control systems must serve diverse user populations across global supply chains. This work, while predating modern statistical software by several decades, established foundational principles for user-friendly statistical analysis that continue to influence the design of quality control and process monitoring tools in manufacturing environments.