Adnan Hassan

Improved SPC chart pattern recognition using statistical features

Increasingly rapid changes and highly precise manufacturing environments require timely monitoring and intervention when deemed necessary. Traditional Statistical Process Control (SPC) charting, a popular monitoring and diagnosis tool, is being improved to be more sensitive to small changes and to include more intelligence to handle dynamic process information. Artificial neural network-based SPC chart pattern recognition schemes have been introduced by several researchers. These schemes need further improvement in terms of generalization and recognition performance. One possible approach is through improvement in data representation using features extracted from raw data. Most of the previous work in intelligent SPC used raw data as input vector representation. The literature reports limited work dealing with features, but it lacks extensive comparative studies to assess the relative performance between the two approaches. The objective of this study was to evaluate the relative performance of a feature-based SPC recognizer compared with the raw data-based recognizer. Extensive simulations were conducted using synthetic data sets. The study focused on recognition of six commonly researched SPC patterns plotted on the Shewhart X-bar chart. The ANN-based SPC pattern recognizer trained using the six selected statistical features resulted in significantly better performance and generalization compared with the raw data-based recognizer. Findings from this study can be used as guidelines in developing better SPC recognition systems.

Issues in quality engineering research

The field of quality has undergone significant changes as reflected by changes in its definition, paradigms, approaches, techniques, and scope of application. This paper reviews emerging trends and issues focusing on quality engineering. Changes in customer expectation have driven the changes in the technology of design and manufacturing, which is becoming more important in satisfying individual customer expectations. This also calls for special attention to the engineering aspects of quality. Brief reviews on recent advances in the prominent quality tools such as statistical process control, quality function deployment, and design of experiment are reported. General trends in quality engineering research show the tools are being enhanced, integrated, computerized and broaden their application bases, where possible opportunities for further investigation are indicated. Among others these include contributions in multiple‐response optimization, intelligent quality systems, multivariate SPC, and practical and simple guidelines for actual implementation of various tools.

A tuned NSGA-II to optimize the total cost and service level for a just-in-time distribution network

Distribution network planning has attracted the attention of many studies during last decades. Just-in-time (JIT) distribution has a key role in efficient delivery of products within distribution networks. In modeling of JIT distribution networks, the most frequently applied objectives are related to cost and service level. However, evaluating the impact of simultaneously minimizing total costs and balance between distribution network entities in different echelons still rarely complies with the current literature. To remedy this shortcoming and model reality more accurately, this paper develops a multi-objective mixed-integer nonlinear optimization model for a JIT distribution in three-echelon distribution network. The aims are minimization of total logistics cost along with maximization of capacity utilization balance for distribution centers and manufacturing plants. A non-dominated sorting genetic algorithm-II (NSGA-II) with three different mutation operators namely swap, reversion and insertion is employed to provide a set of near-optimal Pareto solutions. Then, the provided solutions are verified with non-dominated ranked genetic algorithm (NRGA) as well. The Taguchi method in design of experiments tunes the parameters of both algorithms, and their performances are then compared in terms of some multi-objective performance measures. In addition, a genetic algorithm is used to assess Pareto optimal solutions of NSGA-II. Different problems with different sizes are considered to compare the performance of the suggested algorithms. The results show that the proposed solution approach performs efficiently. Finally, the conclusion and some directions for future research are proposed.

Feature selection for SPC chart pattern recognition using fractional factorial experimental design

Publisher Summary Feature selection is one of the important steps in designing a pattern recognizer. This chapter presents a study to select a minimal set of statistical features for statistical process control (SPC) chart pattern recognition using the fractional factorial experimental design. A resolution IV design was used to identify the significant features from a list of ten possible candidates to represent the input data streams. Further judgment was adopted to arrive at the final selection in the light of some ambiguities among confounded two-factor interactions. The final six selected features set comprising, autocorrelation, cusum, mean, standard deviation, mean-square value, and skewness as the input vector resulted in an average correct classification rate of 97.1% and standard deviation of 0.878. This can be applied to other feature selection problems besides SPC chart pattern recognition.

Carbon-capped Distribution Planning

We use a hybrid NSGA-II to obtain the near Pareto front solutions in a Just-In-Time distribution network.We investigate different carbon constraints namely periodic, cumulative and global.We present the complexity analysis for the proposed model.We analyze the solution approach as well as identify some managerial and policy insights. Products distribution and transportation is one of the largest sources of CO2 emission in supply chains. To date, a number of researchers have argued that intensive transportation activities through popular distribution strategies such as Just-In-Time (JIT) could significantly increase carbon emissions within logistics chains. However, a systematic understanding of how JIT distribution affects carbon emissions is still lacking in current literature. In this study, we develop a bi-objective optimization model for a carbon-capped JIT distribution of multiple products in a multi-period and multi-echelon distribution network. The aims are to jointly minimize total logistics cost and to minimize the maximum carbon quota allowed per period (carbon cap). The considered problem is investigated under three different carbon emission constraints namely periodic, cumulative and global. Since the studied problem is NP-Hard, a non-dominated sorting genetic algorithm-II (NSGA-II) is developed and its parameters are tuned by Taguchi method. For further quality improvement of the developed solution approach, a novel local search approach called modified firefly algorithm incorporates NSGA-II. Different sizes of the problem are considered to compare the performances of the proposed hybrid NSGA-II and the classical one. Finally, the results are presented along with some policy and managerial insights. For policy makers, the findings show the impact of varying the carbon emission cap on total cost and total emissions under JIT distribution concept. From managerial perspectives, we analyze the relationships between average inventory holding and backlog level per period which can assist mangers to identify critical decisions for JIT distribution of products in carbon-capped environment.

A Hybrid Genetic Algorithm with a Knowledge-Based Operator for Solving the Job Shop Scheduling Problems

Scheduling is considered as an important topic in production management and combinatorial optimization in which it ubiquitously exists in most of the real-world applications. The attempts of finding optimal or near optimal solutions for the job shop scheduling problems are deemed important, because they are characterized as highly complex and -hard problems. This paper describes the development of a hybrid genetic algorithm for solving the nonpreemptive job shop scheduling problems with the objective of minimizing makespan. In order to solve the presented problem more effectively, an operation-based representation was used to enable the construction of feasible schedules. In addition, a new knowledge-based operator was designed based on the problem’s characteristics in order to use machines’ idle times to improve the solution quality, and it was developed in the context of function evaluation. A machine based precedence preserving order-based crossover was proposed to generate the offspring. Furthermore, a simulated annealing based neighborhood search technique was used to improve the local exploitation ability of the algorithm and to increase its population diversity. In order to prove the efficiency and effectiveness of the proposed algorithm, numerous benchmarked instances were collected from the Operations Research Library. Computational results of the proposed hybrid genetic algorithm demonstrate its effectiveness.

A literature review on green supply chain modelling for optimising CO 2 emission

Global warming impacts are becoming more visible in our daily life. Supply chain activities and many logistics activities are the leading sources of carbon dioxide (CO2) emission and environmental pollutions. These issues have raised concerns to reduce CO2 emissions amount through design and planning of supply chain networks. Operations research has been recognised by many studies as an effective tool to deal with CO2 emission in design and planning of green supply chains. To date, a number of literature reviews have highlighted the contribution of operations research to green supply chain management with broader areas of focus. In this paper, we present a review which highlights the operations research contribution to recent green supply chain and logistics literature which specifically focuses on planning and control of supply chain activities with respect to CO2 emission. Finally, we propose some possible areas for further developments of current studies and directions for future research.

Synergistic-ANN Recognizers for Monitoring and Diagnosis of Multivariate Process Shift Patterns

An intelligent control chart pattern recognition system is essential for efficient monitoring and diagnosis process variation in automated manufacturing environment. Artificial neural networks (ANN) have been applied for automated recognition of control chart patterns since the last 20 years. In early study, the development of control chart patterns recognizers was mainly based on generalized-ANN model. There has been an increasing trend among researchers to move beyond generalized recognizer particularly for addressing complex recognition tasks. However, the existing works mainly focus on univariate process cases. This paper aims to investigate an effective synergistic-ANN model for on-line monitoring and diagnosis multivariate process patterns. The recognition performances of a generalized-ANN and the parallel distributed ANN recognizers for learning dynamic patterns of multivariate process patterns were discussed.

Toward a dynamic balanced scorecard model for humanitarian relief organizations’ performance management

Purpose In recent years, the balanced scorecard (BSC) has received considerable interest among practitioners for managing their organization’s performance. Unfortunately existing BSC frameworks, particularly for humanitarian supply chains, lack causal relationships among performance indicators, actions, and outcomes. They are not able to provide a dynamic perspective of the organization with factors that drive the organization’s behavior toward its mission. Lack of conceptual references seems to hinder the development of a performance measurement system toward this direction. The paper aims to discuss these issues. Design/methodology/approach The authors formulate the interdependencies among key performance indicators (KPIs) in terms of cause-and-effect relationships based on published case studies reported in international journals from 1996 to 2017. Findings This paper aims to identify the conceptual interdependencies among KPIs and represent them in the form of a conceptual model. Research limitations/implications The study was solely based on relevant existing literature. Therefore further practical research is needed to validate the interdependencies of performance indicators in the strategy map. Practical implications The proposed conceptual model provides the structure of a dynamic balanced scorecard (DBSC) in the humanitarian supply chain and should serve as a starting reference for the development of a practical DBSC model. The conceptual framework proposed in this paper aims to facilitate further research in developing a DBSC for humanitarian organizations (HOs). Originality/value Existing BSC frameworks do not provide a dynamic perspective of the organization. The proposed conceptual framework is a useful reference for further work in developing a DBSC for HOs.

A Framework for Multivariate Process Monitoring and Diagnosis

Monitoring and diagnosis of mean shifts in manufacturing processes become more challenging when involving two or more correlated variables. Unfortunately, most of the existing statistical process control frameworks are only effective in shift detection but suffers high false alarm, that is, imbalanced performance monitoring. The problem becomes more complicated when dealing with small shift particularly in identifying the causable variables. In this research, a framework to address balanced monitoring and accurate diagnosis was investigated. Design considerations involved extensive simulation experiments to select input representation based on raw data and statistical features, recognizer design structure based on synergistic model, and monitoring-diagnosis approach based on two stages technique. The study focuses on correlated process mean shifts for cross correlation function, ρ = 0.1 ~ 0.9 and mean shift, μ = ± 0.75 ~ 3.00 standard deviations. The proposed design, that is, an Integrated Multivariate Exponentially Weighted Moving Average with Artificial Neural Network gave superior performance, namely, average run length, ARL1 = 3.18 ~ 16.75 (for out-of-control process), ARL0 = 452.13 (for in-control process) and recognition accuracy, RA = 89.5 ~ 98.5%. This research has provided a new perspective in realizing balanced monitoring and accurate diagnosis of multivariate correlated process mean shifts.