A highly optimised tolerance-based approach for multi-stage,multi-product supply chain network design

In this paper we propose an algorithm called Highly Optimised Tolerance (HOT) for solving a multi-stage, multi-product supply chain network design problem. HOT is based on power law and control theory. The proposed approach takes its traits from the local incremental algorithm (LIA), which was initially employed to maximise the design parameter (i.e. yield), particularly in the percolation model. The LIA is somewhat analogous to the evolution by natural selection schema. The proposed methodology explores a wide search space and is computationally viable. The HOT algorithm tries to make the system more robust at each step of the optimisation. The objective of this paper is to reduce the total cost of supply chain distribution by selecting the optimum number of facilities in the network. To examine the effectiveness of the HOT algorithm we compare the results with those obtained by applying simulated annealing on a supply chain network design problem with different problem sizes and the same data sets.     

Designing a supply chain network for deteriorating inventory under preservation effort and trade credits

When designing a perishable goods supply chain network under trade credit arrangements, distribution companies must contend with deteriorating inventory and product preservation efforts to maximise profits. Key decisions include locating distribution centres (DCs), assigning retail stores to DCs, joint replenishment cycle time and investing in preservation technology. This paper addresses these factors from the position that as preservation effort increases, preservation technology cost increases and deterioration rate decreases. An algorithm based on piecewise nonlinear optimisation is provided for solving supply chain network design problems efficiently. In contrast to other studies that have used the approximation approach, the proposed approach solves the original problem accurately and efficiently. Numerical studies are conducted to demonstrate the solutions procedures and determine the effects of the parameters on decisions and profits. The results of this study and the proposed modelling approach are useful references for managerial decisions in designing a supply chain network the context of trade credit and inventory deterioration.     

Solving closed-loop supply chain problems using game theoretic particle swarm optimisation

In this paper, we propose a closed-loop supply chain network configuration model and a solution methodology that aim to address several research gaps in the literature. The proposed solution methodology employs a novel metaheuristic algorithm, along with the popular gradient descent search method, to aid location-allocation and pricing-inventory decisions in a two-stage process. In the first stage, we use an improved version of the particle swarm optimisation (PSO) algorithm, which we call improved PSO (IPSO), to solve the location-allocation problem (LAP). The IPSO algorithm is developed by introducing mutation to avoid premature convergence and embedding an evolutionary game-based procedure known as replicator dynamics to increase the rate of convergence. The results obtained through the application of IPSO are used as input in the second stage to solve the inventory-pricing problem. In this stage, we use the gradient descent search method to determine the selling price of new products and the buy-back price of returned products, as well as inventory cycle times for both product types. Numerical evaluations undertaken using problem instances of different scales confirm that the proposed IPSO algorithm performs better than the comparable traditional PSO, simulated annealing (SA) and genetic algorithm (GA) methods.     

Swarm intelligence-based strategic bidding in competitive electricity markets

In the competitive electricity markets, formation of supply bid is one of the main concerns, where suppliers have to maximise their profit under incomplete information of other competing generators. An environment is described in which suppliers bid strategically to sell electricity in a pool market. The bidding decision is optimised from a single supplier's viewpoint in both block-bid and linear-bid models of an electricity market. To include uncertain behaviour of other competing suppliers, two different probabilistic models are used. Their bids are constructed using probability distribution functions obtained from the decision-maker's observations of historical market data. Single supplier's decision-making problem is solved by a modern population-based heuristic algorithm, known as particle swarm optimisation (PSO). Search procedure of PSO is based on the concept of combined effect of cognitive and social learning of the members in a group. The effectiveness of the proposed method is tested with examples and the results are compared with the solutions obtained using the genetic algorithm approach.     

An evolutionary algorithm for a new multi-objective location-inventory model in a distribution network with transportation modes and third-party logistics providers

This paper proposes a multi-objective optimisation algorithm for solving the new multi-objective location-inventory problem (MOLIP) in a distribution centre (DC) network with the presence of different transportation modes and third-party logistics (3PL) providers. 3PL is an external company that performs all or part of a company’s logistics functions. In order to increase the efficiency and responsiveness in a supply chain, it is assumed that 3PL is responsible to manage inventory in DCs and deliver products to customers according to the provided plan. DCs are determined so as to simultaneously minimise three conflicting objectives; namely, total costs, earliness and tardiness, and deterioration rate. In this paper, a non-dominated sorting genetic algorithm (NSGA-II) is proposed to perform high-quality search using two-parallel neighbourhood search procedures for creating initial solutions. The potential of this algorithm is evaluated by its application to the numerical example. Then, the obtained results are analysed and compared with multi-objective simulated annealing (MOSA). It is concluded that this algorithm is capable of generating a set of alternative DCs considering the optimisation of multiple objectives, significantly improving the decision-making process involved in the distribution network design.     

Concurrent optimal allocation of distributed manufacturing resources using extended Teaching-Learning-Based Optimization

The optimal allocation of distributed manufacturing resources is a challenging task for supply chain deployment in the current competitive and dynamic manufacturing environments, and is characterised by multiple objectives including time, cost, quality and risk that require simultaneous considerations. This paper presents an improved variant of the Teaching-Learning-Based Optimisation (TLBO) algorithm to concurrently evaluate, select and sequence the candidate distributed manufacturing resources allocated to subtasks comprising the supply chain, while dealing with the trade-offs among multiple objectives. Several algorithm-specific improvements are suggested to extend the standard form of TLBO algorithm, which is only well suited for the one-dimensional continuous numerical optimisation problem well, to solve the two-dimensional (i.e. both resource selection and resource sequencing) discrete combinatorial optimisation problem for concurrent allocation of distributed manufacturing resources through a focused trade-off within the constrained set of Pareto optimal solutions. The experimental simulation results showed that the proposed approach can obtain a better manufacturing resource allocation plan than the current standard meta-heuristic algorithms such as Genetic Algorithm, Particle Swarm Optimisation and Harmony Search. Moreover, a near optimal resource allocation plan can be obtained with linear algorithmic complexity as the problem scale increases greatly.     

Shape optimisation of a two-phase ejector for CO2 refrigeration systems

The shape optimisation of four CO₂ ejectors is presented in this study. The considered ejectors were originally designed for a multi-ejector supermarket CO₂ refrigeration system. The objective function was formulated to consider the multiple operating regimes, where the goal of the optimisation was to maximise the device efficiency. Six geometrical parameters were considered in the optimisation procedure. The applied optimisation scheme was a combination of a genetic algorithm coupled with the effective and validated CFD tool, ejectorPL. The optimisation results showed that the ejector efficiency improved by 6%. The shape modification trends were similar for all of the considered ejectors. All of the shape modifications resulted in a smoother expansion inside the motive nozzle, less intense turbulence inside the mixing section and a more uniform velocity field inside the mixing section. The obtained results showed that the presented methodology can be effectively used for ejector design for numerous applications.     

An optimisation model for purchase,production and distribution in fish supply chain – a case study

This paper presents an optimisation model for spawn purchase, fish culturing production process and harvested fish distribution in a fish supply chain. Due to the complexity and variety of real-world fish supply chains, the model is built based on a case study for a real trout fish farm to illustrate the methodology on how to incorporate influential factors from both warm chain and cold chain. Warm chain mainly considers the biological factors while fish is alive and cold chain mainly considers the economic factors after fish is ready for harvest, harvested, and processed. The model seeks to improve the trout farm production planning to help decision-making on spawn purchase quantity, the best time to harvest fish, and the farming periods. In addition, the model adopts a customer classification method in distribution planning that is able to prioritise the delivery of fresh fish to the most profitable customers. A mixed integer linear programming (MILP) model was developed to maximise the total profit. The experimental results demonstrate that farmers’ total profit can be increased after applying the proposed optimisation strategy, compared to the traditional farming strategy.     

Towards optimal configuration of a manufacturer’s supply network with demand flexibility

Demand flexibility exhibits the degree to which customers are often willing to compromise on product features or performance levels for budgetary (reflected in price) or schedule (reflected in delivery) reasons. It is essential for a manufacturer to map demand flexibility into the supply side and investigate its impact on supply network configuration to maximise its total profit. This paper is among the first contributions that seek to address the challenge of optimal configuration of a manufacturer’s supply network that consists of raw material suppliers and contract manufacturers, considering demand flexibility and commonality among different product families. A new mixed integer programming model is developed to describe the characteristics of this problem. The objective was to maximise the manufacturer’s total profit subject to various operating constraints of the supply chain. In view of the complexity and non-deterministic polynomial-time (NP)-hard nature of the problem, a hybrid constraint programming and simulated annealing algorithm is proposed to solve the problem optimally. Extensive numerical studies are conducted to validate the effectiveness of the proposed model and the hybrid algorithm.     

A particle swarm optimisation for the no-wait flow shop problem with due date constraints

This paper considers the no-wait flow shop scheduling problem with due date constraints. In the no-wait flow shop problem, waiting time is not allowed between successive operations of jobs. Moreover, a due date is associated with the completion of each job. The considered objective function is makespan. This problem is proved to be strongly NP-Hard. In this paper, a particle swarm optimisation (PSO) is developed to deal with the problem. Moreover, the effect of some dispatching rules for generating initial solutions are studied. A Taguchi-based design of experience approach has been followed to determine the effect of the different values of the parameters on the performance of the algorithm. To evaluate the performance of the proposed PSO, a large number of benchmark problems are selected from the literature and solved with different due date and penalty settings. Computational results confirm that the proposed PSO is efficient and competitive; the developed framework is able to improve many of the best-known solutions of the test problems available in the literature.     

Automated setup planning in CAPP: a modified particle swarm optimisation-based approach

Comprehensive process planning is the key technology for linking design and the manufacturing process and is a rather complex and difficult task. Setup planning has a basic role in computer-aided process planning (CAPP) and significantly affects the overall cost and quality of machined parts. This paper presents a generative system and particle swarm optimisation algorithm (PSO) approach to the setup planning of a given part. The proposed approach and optimisation methodology analyses constraints such as the TAD (tool approach direction), the tolerance relation between features and feature precedence relations, to generate all possible process plans using the workshop resource database. Tolerance relation analysis has a significant impact on setup planning to obtain part accuracy. Based on technological constraints, the PSO algorithm approach, which adopts the feature-based representation, optimises the setup planning using cost indices. To avoid becoming trapped in local optima and to explore the search space extensively, several new operators have been developed to improve the particles’ movements, combined into a modified PSO algorithm. A practical case study is illustrated to demonstrate the effectiveness of the algorithm in optimising the setup planning.     

Multi-objective imperfect preventive maintenance optimisation with NSGA-II

Maintenance optimisation is a multi-objective problem in nature, and it usually needs to achieve a trade-off among the conflicting objectives. In this study, a multi-objective maintenance optimisation (MOMO) model is proposed for electromechanical products, where both the soft failure and hard failure are considered, and minimal repair is performed accordingly. Imperfect preventive maintenance (IPM) is carried out during the preplanned periods, and modelled with a hybrid failure rate model and quasi-renewal coefficient. The initial IPM period and the total number of IPM periods are set as the decision variables, and a MOMO model is developed to optimise the availability and cost rate concurrently. The fast elitist non-dominated sorting genetic algorithm (NSGA-II) is applied to solve the model. A case study of wind turbine’s gearbox is provided. The results show that there are 30 optimal solutions in the MOMO’s Pareto frontier that can maximise the availability and minimise the cost rate simultaneously. Compared with the single-objective maintenance optimisation, it can provide more choices for maintenance decision, and better satisfy the resource constraints and the customer’s preference. The results of the sensitivity analysis show that the effect of age reduction factor on optimisation results is greater than that of failure rate increase factor.     

A dynamic programming approach to integrated assembly planning and supplier assignment with lead time constraints

As manufacturers face fierce competition in the global market, responsiveness has become an important competitiveness factor in addition to quality and cost. One essential responsiveness strategy is to reduce product development and lead times by integrating assembly planning with supplier assignment. This paper addresses the problem of integrated assembly and supply chain design under lead-time constraints by formulating and solving an optimisation problem with minimal total supply chain costs. This new time-constrained joint optimisation problem belongs to an NP-hard resource-constrained scheduling problem. To model this problem effectively, we develop a novel Hyper AND/OR graph and apply it for integrating assembly and supply chain decisions. We also develop a dynamic programming model and associated algorithm in order to solve the integrated optimisation problem with pseudo-polynomial time complexity in practice. Numerical case studies validate that the methods developed can solve the integrated decision-making problem optimally and efficiently. This paper overcomes the limitations of previous studies on concurrent assembly decomposition and supplier selection, which optimises cost without time constraints. The models and results of this research can be applied to a variety of areas including assembly design, maintenance module planning and supply chain restructuring.     

Genetic-algorithms-based algorithm portfolio for inventory routing problem with stochastic demand

This paper presents an algorithm portfolio methodology based on evolutionary algorithms to solve complex dynamic optimisation problems. These problems are known to have computationally complex objective functions, which make their solutions computationally hard to find, when problem instances of large dimensions are considered. This is due to the inability of the algorithms to provide an optimal or near-optimal solution within an allocated time interval. Therefore, this paper employs a bundle of evolutionary algorithms (EAs) tied together with several processors, known as an algorithm portfolio, to solve a complex optimisation problem such as the inventory routing problem (IRP) with stochastic demands. EAs considered for algorithm portfolios are the genetic algorithm and its four variants such as the memetic algorithm, genetic algorithm with chromosome differentiation, age-genetic algorithm, and gender-specific genetic algorithm. In order to illustrate the applicability of the proposed methodology, a generic method for algorithm portfolios design, evaluation, and analysis is discussed in detail. Experiments were performed on varying dimensions of IRP instances to validate different properties of algorithm portfolio. A case study was conducted to illustrate that the set of EAs allocated to a certain number of processors performed better than their individual counterparts.     

A hybrid particle swarm based method for process planning optimisation

A process planning (PP) problem is defined as to determine a set of operation-methods (machine, tool, and set-up configuration) that can convert the given stock to the designed part. Essentially, the PP problem involves the simultaneous decision making of two tasks: operation-method selection and sequencing. This is a combinatorial optimisation problem and it is difficult to find the best solution in a reasonable amount of time. In this article, an optimisation approach based on particle swarm optimisation (PSO) is proposed to solve the PP problem. Due to the characteristic of discrete process planning solution space and the continuous nature of the original PSO, a novel solution representation scheme is introduced for the application of PSO in solving the PP problem. Moreover, two kinds of local search algorithms are incorporated and interweaved with PSO evolution to improve the best solution in each generation. The numerical experiments and analysis have demonstrated that the proposed algorithm is capable of gaining a good quality solution in an efficient way.     

Dynamic co-ordinated scheduling in the supply chain under a process modernisation

A new dynamic model for co-ordinated scheduling of interlinked processes in a supply chain under a process modernisation is presented. Such a problem is vital in many of the supply chain management domains. This problem is represented as a special case of the scheduling problem with dynamically distributed jobs. The peculiarity of the proposed approach is the dynamic interpretation of scheduling based on a natural dynamic decomposition of the problem and its solution with the help of a modified form of continuous maximum principle blended with combinatorial optimisation. The special properties of the developed model allow using methods of discrete optimisation for the schedule calculation. Optimality and sufficiency conditions as well as structural properties of the model are investigated. Advantages and limitations of the proposed approach are discussed. With the developed approach, an explicit inclusion of a process modernisation in the SC co-ordinated decisions for a wide ranges of possible applications as well as a dynamic model and a tractable algorithm for optimal discrete time scheduling on the basis of continuous maximum principle have been obtained.     

Resilient and sustainable supply chain design: sustainability analysis under disruption risks

Resilience to disruptions and sustainability are both of paramount importance to supply chains. However, the interactions between the two have not been thoroughly explored in the academic literature. We attempt to contribute to this area by presenting a hybrid methodology for the design of a sustainable supply network that performs resiliently in the face of random disruptions. A stochastic bi-objective optimisation model is developed that utilises a fuzzy c-means clustering method to quantify and assess the sustainability performance of the suppliers. The proposed model determines outsourcing decisions and resilience strategies that minimise the expected total cost and maximise the overall sustainability performance in disruptions. Important managerial insights and practical implications are obtained from the model implementation in a case study of plastic pipe industry.     

Ant colony optimisation algorithm for distribution-allocation problem in a two-stage supply chain with a fixed transportation charge

In a fixed charge transportation problem, each route is associated with a fixed charge (or a fixed cost) and a transportation cost per unit transported. The presence of the fixed cost makes the problem difficult to solve, thereby requiring the use of heuristic methods. In this paper, an algorithm based on ant colony optimisation is proposed to solve the distribution-allocation problem in a two-stage supply chain with a fixed transportation cost for a route. A numerical study on benchmark problem instances has been carried out. The results obtained for the proposed algorithm have been compared with that for the genetic algorithm-based heuristic currently available in the literature. It is statistically confirmed that the proposed algorithm provides significantly better solutions.     

Optimum part deposition orientation for multiple objectives in SL and SLS prototyping

In the present work an attempt has been made to achieve minimum average part surface roughness (best overall surface quality), minimum build time and support structure for stereolithography (SL) and selective laser sintering (SLS) processed parts by determining optimum part deposition orientation. A conventional optimisation algorithm based on a trust region method (available with MATLAB-7 optimisation tool box) has been used to solve the multi-objective optimisation problem. It is observed that the problem is highly multi-modal in nature and a suitable initial guess, which is used as an input to execute the optimisation module, is important to achieve a global optimum. A simple methodology has been proposed to find out the initial guess so that global minimum is obtained. Finally the surface roughness simulation is carried out with optimum part deposition orientation to have an idea of surface roughness variation over the entire part's surface before depositing the part. Case studies are presented to demonstrate the capabilities of the developed system. The major achievements of this work are consideration of multiple objectives for the two rapid prototyping processes, successful use of conventional optimisation algorithm available with MATLAB to handle multiple objectives and development of graphical user interface-based system.     

An improved ant colony optimisation algorithm for three-tier supply chain scheduling based on networked manufacturing

This paper presents a study on supply chain scheduling from the perspective of networked manufacturing (NM). According to feature analysis of supply chain scheduling based on NM, we comprehensively consider the combined benefits of cost, time, and satisfaction level for customised services. In order to derive a scheduling strategy among supply chain members based on NM, we formulate a three-tier supply chain scheduling model composed of manufacturer, collaborative design enterprise and customer. Three objective functions – time function, cost function and delay punishment function – are employed for model development. We also take into account multi-objective optimisation under the constraint of product capacity. By using an improved ant colony optimisation algorithm, we add different pheromone concentrations to selected nodes that are obtained from feasible solutions and we confine pheromone concentrations τ within the minimum value τ _min and the maximum value τ _max, thus obtaining optimal results. The results obtained by applying the proposed algorithm to a real-life example show that the presented scheduling optimisation algorithm has better convergence, efficiency, and stability than conventional ant colony optimisation. In addition, by comparing with other methods, the output results indicate that the proposed algorithm also has better solutions.