Rule-Based Scheduling for Hot-Rolled Steel Pipes

Literature Overview

This paper by Li Jianxiang, Tang Lixin, Wu Huijiang, and Pang Hali from the School of Information Science and Engineering at Northeastern University, published in "Iron and Steel" (Vol. 39, No. 9, 2004, pp. 39-42), addresses the production scheduling problem for hot-rolled steel pipes. The authors summarize field experience and fuzzy rules from pipe mill operations, formalize them into seven clear rules for batch grouping and sequencing, and propose a novel heuristic method that simultaneously performs batch grouping and sequencing. This approach differs from previously published methods that require batch grouping to be completed before sequencing. The work was supported by the National Natural Science Foundation of China (Grant No. 70171030) and the Fok Ying Tung Education Foundation (Grant No. 81073).

Scheduling Rules and Heuristic Methodology

The core contribution of this paper lies in the systematic formalization of scheduling knowledge that previously existed only as tacit experience among mill operators. The seven rules extracted from field practice govern two primary decisions: which orders to group into a single rolling batch, and in what sequence to process orders within a batch. The following table summarizes the key scheduling considerations:

The heuristic method proposed by the authors operates on the principle that batch grouping and sequencing are interdependent decisions that should be solved simultaneously rather than sequentially. In traditional approaches, batch grouping is first completed based on similarity criteria, and then each batch is sequenced independently. The proposed method recognizes that the optimal sequencing within a batch may influence which orders should be grouped together, and vice versa. This coupled approach enables a more globally optimal solution.

The case study presented demonstrates the practical application of the method to a real production scenario at a steel pipe mill. The results show improvements in roll change frequency reduction, trim loss minimization, and on-time delivery performance compared to conventional scheduling approaches.

Technical Analysis of the Scheduling Problem

From a process engineering perspective, the hot-rolled steel pipe scheduling problem involves several unique challenges that distinguish it from general manufacturing scheduling. First, the rolling mill has a finite number of roll sets, each designed for specific diameter and wall thickness ranges. Changing roll sets is time-consuming and expensive, so minimizing roll changes is a primary objective. Second, the furnace temperature must be carefully controlled for each steel grade, and rapid temperature changes can degrade roll life and product quality. Third, the finishing operations—cutting, straightening, and testing—must be coordinated with the rolling schedule to prevent bottlenecks and excessive inventory.

The formalization of fuzzy rules into clear, actionable criteria represents a significant methodological contribution. In many manufacturing environments, scheduling decisions rely heavily on operator intuition, which can be inconsistent and difficult to replicate. By extracting and codifying these rules, the authors enable more systematic and repeatable scheduling decisions that can be implemented in computer-based scheduling systems.

Integration with Engineering Practice

In practice, the scheduling of hot-rolled steel pipe production involves coordination across multiple departments including sales, production planning, rolling mill operations, and quality assurance. The rules identified in this paper directly address the interface between commercial requirements (delivery dates, order priorities) and technical constraints (roll availability, furnace capacity, material compatibility). For engineers responsible for production planning, understanding these rules provides a structured framework for making scheduling decisions that balance competing objectives.

The simultaneous batch grouping and sequencing approach is particularly relevant for mills operating with limited roll sets and high order variability. In such environments, the sequential approach often leads to suboptimal solutions where batches are well-formed but poorly sequenced, or vice versa. The coupled method enables a more holistic optimization that considers both intra-batch and inter-batch interactions.

Study Insights and Practical Implications

This paper represents an important contribution to the body of knowledge on steel pipe production scheduling. The systematic extraction of expert knowledge and its formalization into computable rules bridges the gap between operational experience and algorithmic implementation. For engineers in the steel pipe industry, the practical value lies in the recognition that scheduling optimization is not merely a computational exercise but requires deep understanding of the physical constraints and quality implications of each scheduling decision. The seven rules provide a practical checklist for evaluating scheduling proposals, and the heuristic method offers a viable approach for real-time scheduling in production environments where computational resources and time are limited. The methodology demonstrates that combining domain-specific knowledge with algorithmic innovation can yield meaningful improvements in manufacturing efficiency without requiring prohibitively expensive computational infrastructure.