Srikanta Routroy

Comparing the performance of neural networks developed by using Levenberg-Marquardt and Quasi-Newton with the gradient descent algorithm for modelling a multiple response grinding process

Highlights? Levenberg-Marquardt (L-M) and Boyden, Fletcher, Goldfarb and Shanno (BFGS) update Quasi-Newton (Q-N)-based BPNN networks are equally efficient as adaptive learning (A-L) algorithm-based BPNN network. ? L-M algorithm has fastest network convergence rate, followed by BFGS update Q-N and A-L algorithm. ? A-L -based BPNN learns faster than BFGS update Q-N, and L-M takes maximum time for network training. ? A-L algorithm is relatively easy-to-understand and implement, as compared to L-M or BFGS update Q-N algorithm, for online process control. Monitoring and control of multiple process quality characteristics (responses) in grinding plays a critical role in precision parts manufacturing industries. Precise and accurate mathematical modelling of multiple response process behaviour holds the key for a better quality product with minimum variability in the process. Artificial neural network (ANN)-based nonlinear grinding process model using backpropagation weight adjustment algorithm (BPNN) is used extensively by researchers and practitioners. However, suitability and systematic approach to implement Levenberg-Marquardt (L-M) and Boyden, Fletcher, Goldfarb and Shanno (BFGS) update Quasi-Newton (Q-N) algorithm for modelling and control of grinding process is seldom explored. This paper provides L-M and BFGS algorithm-based BPNN models for grinding process, and verified their effectiveness by using a real life industrial situation. Based on the real life data, the performance of L-M and BFGS update Q-N are compared with an adaptive learning (A-L) and gradient descent algorithm-based BPNN model. The results clearly indicate that L-M and BFGS-based networks converge faster and can predict the nonlinear behaviour of multiple response grinding process with same level of accuracy as A-L based network.

Evaluating the critical success factors of supplier development: a case study

Purpose – The purpose of this paper is to identify and evaluate the critical success factors (CSFs) responsible for supplier development (SD) in a manufacturing supply chain environment.Design/methodology/approach – In total, 13 CSFs for SD are identified (i.e. long‐term strategic goal; top management commitment; incentives; suppliers supplier condition; proximity to manufacturing base; supplier certification; innovation capability; information sharing; environmental readiness; external environment; project completion experience; supplier status and direct involvement) through extensive literature review and discussion held with managers/engineers in different Indian manufacturing companies. A fuzzy analytic hierarchy process (FAHP) is proposed and developed to evaluate the degree of impact of each CSF on SD.Findings – The degree of impact for each CSF on SD is established for an Indian company. The results are discussed in detail with managerial implications. The long‐term strategic goal is found to be ...

Differential evolution algorithm for supply chain inventory planning

Purpose – This paper discusses the inventory planning of a supply chain, which consists of a manufacturer, distributor and retailer.Design/methodology/approach – The differential evolution algorithm is developed to minimize the total system‐wide cost, which consists of supply chain inventory capital, supply chain ordering/set‐up cost and supply chain stock‐out cost.Findings – The differential evolution algorithm helps in determining ordering/production quantity and inventory/service level that should be maintained by each member of the supply chain.Originality/value – The algorithm developed is useful in increasing the customer service level and in decreasing the inventory level throughout the supply chain.

Analyzing the performance of supplier development: a case study

Purpose – The first objective of this paper is to identify the critical success factors (CSFs) and their corresponding key performance indicators (KPIs) for supplier development (SD) in a manufacturing environment. The second objective is to develop a methodology to analyze and evaluate the performance for SD using CSFs and their KPIs over the time. Design/methodology/approach – In all, 13 CSFs and their corresponding KPIs for SD are established through an extensive literature review, discussion held with managers/engineers in different Indian manufacturing companies and conducting brainstorming sessions. A methodology is proposed using analytic hierarchy process (AHP) and performance value analysis to assess and evaluate the performance of SD over the time. Findings – From an extensive analysis of the results, under the given circumstances, the growth of SD performance is positive at different progressive points along the time horizon. Research limitations/implications – This study has not been statistic...

Analyzing supplier development program enablers using fuzzy DEMATEL

Purpose – The purpose of this paper is to identify, quantify and establish relationship (i.e. cause and effect) among various supplier development program enablers (SDPEs) in a specific manufacturing environment. Design/methodology/approach – The proposed methodology runs into four phases, i.e. defining supplier development program (SDP) environment, identifying relevant SDPEs, collecting experts’ qualitative opinions regarding SDPEs and analyzing the SDPEs using Fuzzy DEMATEL (Decision Making Trial and Evaluation Laboratory). The fourth phase is programmed using Matlab 7.10.0 (R2010a). The proposed methodology is implemented in an Indian manufacturing company and the results are analyzed to provide directions for the company while implementing SDPs. Findings – The proposed methodology leads to the ranking of SDPEs, classification of SDPEs into cause and effect groups and establishment of interactions for each SDPE using impact relationship map. Of the 20 SDPEs considered in the analysis, “top management ...

DECISION FRAMEWORK FOR SELECTION OF FACILITIES LOCATION IN COMPETITIVE SUPPLY CHAIN

Supply chain management (SCM) is an area that has recently received a great deal of attention. In todays markets, no business can be successful without mastering the issues, problems and possibilities in managing competitive supply chains. In competitive supply chain network, facilities location includes location of manufacturing plant, which is considered as supply chain design decisions and involved in major capital investments and has a long-term effect on the supply chain performance. Therefore, the successful execution of this decision would give cutting edge to the organization. Selection of a facilities location is a complex task as it involves both qualitative and quantitative factors. The present work describes a framework for selection of facilities location in competitive supply chain.

Framework for green procurement: a case study

Manufacturing companies are putting significant effort and investment to achieve feasible and affordable greenness in their supply chains. The procurement is a major area of manufacturing supply chain and has a significant impact on the final product with respect to quality, cost and greenness. Therefore, the manufacturing organisation should incorporate green issues while procuring components, parts, sub-assembly and services from suppliers. In this paper, a conceptual framework for green procurement (GP) in manufacturing environment is proposed. The simplicity and clarity of proposed model enhances its acceptability for implementation and also ensures desired level of greenness in the procurement process.

Measurement of manufacturing agility: a case study

Purpose – The purpose of this paper is to determine the agility level of a manufacturing system along different timelines. Design/methodology/approach – The fuzzy synthetic extents of agile manufacturing enablers (AMEs), on the basis of their importance, are determined. Then they are integrated with their performance ratings along different timeline for calculating the Fuzzy Agile Manufacturing Index (FAMI). Euclidean distances of FAMI from predetermined agility levels are mapped to determine the agility level of the manufacturing system along different timeline. Findings – The proposed methodology was implemented in an Indian manufacturing organization to determine its agility level. It was concluded from the obtained results that there was significant improvement in the agility level along the timeline. Research limitations/implications – The weights of the AMEs are assumed to be constant along the timeline. Practical implications – The supply chain mangers can easily apply this methodology in their res...

Benchmarking model of supplier development for an Indian gear manufacturing company

Purpose – The aim of this paper is to propose a benchmarking model of supplier development (SD) for an Indian gear manufacturing company for its successful adoption and improvement in a continuous basis. Design/methodology/approach – Thirteen SD critical success factors (CSFs) are identified and classified into four categories through extensive literature review and discussion held with managers/senior engineers in different Indian manufacturing companies. The four categories are primary CSFs related to supplier, secondary CSFs related to supplier, CSFs related to manufacturer and CSFs related to both manufacturer and supplier. The interpretative structural modeling (ISM) approach is applied to the Indian gear manufacturing company for developing and analyzing structural framework of CSFs to propose a benchmarking model for SD. Findings – It is concluded that SD adoption should be carried out in four phases sequentially for the Indian gear manufacturing company. The first, second, third and fourth phase s...

Selection of Third Party Logistics Provider in supply chain

For most of the manufacturing firms, logistics is not considered to be a core competency and it is often performed by Third Party Logistics Provider (TPLP). The present work describes an efficient decision framework for TPLP evaluation and selection in supply chain. The objective of this paper is to explain how an Analytic Hierarchy Process (AHP) and Performance Value Analysis (PVA) algorithm can be used to capture and analyse Significant Categories (SCs) and Performance Indicators (PIs) for ranking the TPLPs effectively. The application of decision framework for TPLP evaluation and selection has been demonstrated with a case situation.