Template-Type: ReDIF-Article 1.0
Author-Name: Marialuisa Menanno
Author-X-Name-First: Marialuisa
Author-X-Name-Last: Menanno
Author-Name: Matteo Mario Savino
Author-X-Name-First: Matteo Mario
Author-X-Name-Last: Savino
Author-Name: Filippo Emanuele Ciarapica
Author-X-Name-First: Filippo Emanuele
Author-X-Name-Last: Ciarapica
Title: Exploring continuous improvement for safety management systems through artificial neural networks
Abstract:
The present work investigates safety assessment in Safety Management Systems with the twofold objective of (i) developing safety audits with objective definition of risk levels for workers' activities and (ii) concurrent prioritising the corrective actions. The work is developed within a firm producing automotive components, where safety audits have been managed through a management approach able to (i) define the risk factors for each activity and (ii) quantify the corresponding risk level that may require corrective actions. Risk prediction matrices have been conceived to assess the risk levels. Then, each matrix has been the training set of an Artificial Neural Network used to quantify the values of risk levels. The findings of this study provided some general principles to integrate safety assessment within the continuous improvement and some outcome relative to the use of artificial intelligence for safety assessments.
Journal: Int. J. of Product Development
Pages: 213-241
Issue: 3
Volume: 25
Year: 2021
Keywords: industrial safety; risk assessment; continuous improvement; artificial neural networks.
File-URL: http://www.inderscience.com/link.php?id=118040
File-Format: text/html
File-Restriction: Access to full text is restricted to subscribers.
Handle: RePEc:ids:ijpdev:v:25:y:2021:i:3:p:213-241

Template-Type: ReDIF-Article 1.0
Author-Name: Shubhi Agarwal
Author-X-Name-First: Shubhi
Author-X-Name-Last: Agarwal
Author-Name: Anand Mohan Agrawal
Author-X-Name-First: Anand Mohan
Author-X-Name-Last: Agrawal
Title: Measuring relationship quality in luxury product: exploratory factor analysis
Abstract:
This paper is focused on understanding the perception of the customers towards the quality of their bond with their respective sellers in the Indian jewellery sector and identifying the key factors playing a major role in measuring their relationship quality. For data collection, 407 regular customers of different jewellery showrooms were asked to mark their responses in questionnaire about relationship quality efforts from the retailer's side on a Likert scale. Exploratory factor analysis was used to extract the variables from the data. This research finds out five factors for measuring relationship. 'Commitment', 'Trust', 'Product category involvement', 'Satisfaction' and 'Communication' are the factors that explain the quality of relationships in a luxury product. The focus of the customers is now on systematic, transparent, trustworthy and empathetic dealing. To the best knowledge of the authors, no study has focused on the Indian jewellery sector in the relationship quality context.
Journal: Int. J. of Product Development
Pages: 242-254
Issue: 3
Volume: 25
Year: 2021
Keywords: relationship quality; exploratory factor analysis; jewellery; high-involvement category products; luxury products.
File-URL: http://www.inderscience.com/link.php?id=118043
File-Format: text/html
File-Restriction: Access to full text is restricted to subscribers.
Handle: RePEc:ids:ijpdev:v:25:y:2021:i:3:p:242-254

Template-Type: ReDIF-Article 1.0
Author-Name: Samuele Marinello
Author-X-Name-First: Samuele
Author-X-Name-Last: Marinello
Author-Name: Francesco Lolli
Author-X-Name-First: Francesco
Author-X-Name-Last: Lolli
Author-Name: Rita Gamberini
Author-X-Name-First: Rita
Author-X-Name-Last: Gamberini
Title: An open innovation B2B web platform design: application of the QFD approach for the definition of its primary functions
Abstract:
Global markets and the concept of innovation require modern companies to quickly adapt to two very relevant paradigms: digital innovation and open innovation. Therefore, the use of digital technologies and the development of open collaboration networks have radically changed the nature of the organisation of innovation and of the managerial approach and strategic choices. The objective of this paper is to describe the approach adopted to define the main functionalities of a digital Business to Business (B2B) platform for the development of new commercial collaborations between Small and Medium-sized Enterprises (SMEs). The approach of Quality Function Deployment and its House of Quality tool have been applied to support the combination of customer and technical needs. The prioritisation of the technical characteristics of the platform has identified in the 'system for managing orders' and 'systems to speed up processes' the main functions to be developed with greater attention within the platform.
Journal: Int. J. of Product Development
Pages: 255-282
Issue: 3
Volume: 25
Year: 2021
Keywords: open innovation; digital innovation; web platform; quality function deployment; house of quality.
File-URL: http://www.inderscience.com/link.php?id=118045
File-Format: text/html
File-Restriction: Access to full text is restricted to subscribers.
Handle: RePEc:ids:ijpdev:v:25:y:2021:i:3:p:255-282

Template-Type: ReDIF-Article 1.0
Author-Name: Hongli Wan
Author-X-Name-First: Hongli
Author-X-Name-Last: Wan
Author-Name: Yuchen Li
Author-X-Name-First: Yuchen
Author-X-Name-Last: Li
Title: A personalised recommendation algorithm of user preference products based on Bayesian network
Abstract:
In order to overcome the problems of low recommendation accuracy, coverage rate and user diversity index in current personalised recommendation algorithms for user preference products, a new personalised recommendation algorithm based on Bayesian network is proposed. The algorithm takes into account the changing rule of users' interest characteristics with time, and divides the friendly neighbour network. The tags that users are interested in are obtained by user tag information and network partition results, the user's preference for products is obtained by combining with Bayesian network, and personalised products are recommended for users according to the results of preference calculation. The simulation results show that the proposed algorithm can effectively increase the accuracy, coverage and diversity index of user preference products, and recommend the most satisfactory products for users.
Journal: Int. J. of Product Development
Pages: 85-100
Issue: 2
Volume: 25
Year: 2021
Keywords: Bayesian network; user preference; product; personalised recommendation.
File-URL: http://www.inderscience.com/link.php?id=116147
File-Format: text/html
File-Restriction: Access to full text is restricted to subscribers.
Handle: RePEc:ids:ijpdev:v:25:y:2021:i:2:p:85-100

Template-Type: ReDIF-Article 1.0
Author-Name: Hui Zhou
Author-X-Name-First: Hui
Author-X-Name-Last: Zhou
Title: Error correction method of enterprise product cost accounting based on machine learning algorithm
Abstract:
In order to solve the problems of large error correction result and long-time consumption in traditional cost accounting methods, an error correction method of enterprise product cost accounting based on machine learning algorithm is proposed. By analysing the specific cost chain and accounting error sources of enterprise product production, the influence of uncontrollable factors on accounting error is preliminarily avoided. On this basis, the total error of cost accounting is determined by constructing error measurement model. Finally, the optimal solution of error calculation is obtained by using the support vector machine concept in machine learning algorithm, so as to realise error correction. The experimental results show that the correction time consumption of the proposed method is short, the accuracy is high, and the memory space is small, which indicates that the method has a certain application value and can be widely used in the field of enterprise cost accounting.
Journal: Int. J. of Product Development
Pages: 101-113
Issue: 2
Volume: 25
Year: 2021
Keywords: machine learning; cost accounting; error correction; fuzzy support vector machine; cost chain; error measurement model.
File-URL: http://www.inderscience.com/link.php?id=116148
File-Format: text/html
File-Restriction: Access to full text is restricted to subscribers.
Handle: RePEc:ids:ijpdev:v:25:y:2021:i:2:p:101-113

Template-Type: ReDIF-Article 1.0
Author-Name: Yang Wang
Author-X-Name-First: Yang
Author-X-Name-Last: Wang
Title: Risk prediction of information leakage in new product development stage based on data driven model
Abstract:
In order to overcome the problem of low prediction accuracy of current information disclosure risk prediction methods, this paper proposes a new risk prediction method for information leakage based on data-driven model. Based on the data-driven model, the new product development implementation process model is established. The generation and transmission characteristics of information in the new product development stage are analysed. The relationship between the information owner and the information thief is analysed by using the game model. According to the analysis results, the information leakage risk evaluation index system is constructed, and the comprehensive fuzzy evaluation method is used to solve the information leakage. According to the evaluation results, the information leakage risk is predicted. The experimental results show that the index significance coefficient of the proposed method is high, the critical ratio value can be controlled within 2, and the prediction accuracy is high.
Journal: Int. J. of Product Development
Pages: 114-129
Issue: 2
Volume: 25
Year: 2021
Keywords: data driven model; new product development; information leakage; game model; risk prediction.
File-URL: http://www.inderscience.com/link.php?id=116149
File-Format: text/html
File-Restriction: Access to full text is restricted to subscribers.
Handle: RePEc:ids:ijpdev:v:25:y:2021:i:2:p:114-129

Template-Type: ReDIF-Article 1.0
Author-Name: Yang He
Author-X-Name-First: Yang
Author-X-Name-Last: He
Title: Automatic adjustment method of sports equipment parameters based on dynamic data drive
Abstract:
In order to overcome the problems of low sensitivity, large adjustment deviation and low adjustment efficiency existing in the traditional parameter adjustment method of sports fitness equipment, this paper proposes a new automatic parameter adjustment method of squatting fitness equipment based on dynamic data drive. This method analyses the basic elements of sports fitness equipment, and obtains the relevant parameters of squatting fitness equipment. A multi-objective optimisation model is established to optimise the support probability and utilisation rate of spare parts. According to the optimisation results, Kalman filter method is used to fuse the equipment parameters, and the method of random gradient descent and dynamic data-driven is used to adjust the fused parameters. The experimental results show that it has obvious advantages in equipment sensitivity, parameter adjustment deviation and adjustment efficiency. After adjustment, the highest sensitivity of equipment is close to 0.9.
Journal: Int. J. of Product Development
Pages: 130-144
Issue: 2
Volume: 25
Year: 2021
Keywords: dynamic data drive; sports equipment; parameter adjustment; Kalman filter; multi-objective optimisation model; dynamic data drive.
File-URL: http://www.inderscience.com/link.php?id=116150
File-Format: text/html
File-Restriction: Access to full text is restricted to subscribers.
Handle: RePEc:ids:ijpdev:v:25:y:2021:i:2:p:130-144

Template-Type: ReDIF-Article 1.0
Author-Name: Hongtao Zhang
Author-X-Name-First: Hongtao
Author-X-Name-Last: Zhang
Title: Resource sharing method of new product development in the digital market based on blockchain
Abstract:
In order to overcome the problems of long query time and low efficiency in the process of new product development resource sharing in digital market, this paper proposes a new resource sharing method for new product development in digital market based on blockchain. Based on the blockchain technology, this method constructs the knowledge sharing system of new product development in digital market. This paper analyses the faults existing in the system shared data, and introduces the contents of these faults in the technical part into ontology, constructs the knowledge sharing model, promotes the resource sharing mechanism of new product development, and completes the resource integration and sharing of new product development according to the ontology knowledge base. The experimental results show that the query time is controlled within 80 s, and the time efficiency is higher.
Journal: Int. J. of Product Development
Pages: 145-159
Issue: 2
Volume: 25
Year: 2021
Keywords: blockchain technology; digital market; new product development; resource sharing.
File-URL: http://www.inderscience.com/link.php?id=116151
File-Format: text/html
File-Restriction: Access to full text is restricted to subscribers.
Handle: RePEc:ids:ijpdev:v:25:y:2021:i:2:p:145-159

Template-Type: ReDIF-Article 1.0
Author-Name: Zhe Sun
Author-X-Name-First: Zhe
Author-X-Name-Last: Sun
Title: Performance evaluation method of social public sports goods development based on fuzzy comprehensive evaluation
Abstract:
In order to overcome the problems of low accuracy and long time-consuming of traditional performance evaluation methods, this paper proposes a new performance evaluation method of social public sports goods development based on fuzzy comprehensive evaluation. The EEP-LCA evaluation system is established based on the development process and evaluation objectives of social public sports goods. The fuzzy comprehensive evaluation method is introduced to determine the factor set and weight set, so as to obtain the weight of the first level evaluation index and the second level evaluation index. According to the weight calculation results and the principle of fuzzy vector single value, the comprehensive evaluation results are obtained. The experimental results show that the highest fitting degree between the actual results and the evaluation results is 85%, and the evaluation time is less than 7 s, and the evaluation accuracy is high.
Journal: Int. J. of Product Development
Pages: 160-174
Issue: 2
Volume: 25
Year: 2021
Keywords: fuzzy comprehensive evaluation; social public sports supplies; performance evaluation; EEP-LCA evaluation system; factor set; weight set.
File-URL: http://www.inderscience.com/link.php?id=116152
File-Format: text/html
File-Restriction: Access to full text is restricted to subscribers.
Handle: RePEc:ids:ijpdev:v:25:y:2021:i:2:p:160-174

Template-Type: ReDIF-Article 1.0
Author-Name: Zhiming Zhong
Author-X-Name-First: Zhiming
Author-X-Name-Last: Zhong
Title: Optimisation of data transmission delay of heart rate monitoring equipment in sports
Abstract:
In order to overcome the problem of high transmission delay of traditional heart rate monitoring equipment, this paper proposes a new optimisation method for data transmission delay of sports heart rate monitoring equipment. This paper analyses the characteristics of heart rate data under different exercise intensity, constructs the energy consumption model of data transmission, and analyses the operation process of energy consumption model in detail. According to the RBF neural network, the related parameters affecting the data transmission delay of heart rate monitoring equipment are calculated and optimised to complete the optimisation of data transmission delay of heart rate monitoring equipment. The experimental results show that the data delay optimisation method proposed in this paper can effectively reduce the transmission delay, the minimum high delay time is only 60 ms, improve the transmission rate, and has strong practical application.
Journal: Int. J. of Product Development
Pages: 175-186
Issue: 2
Volume: 25
Year: 2021
Keywords: sports; heart rate monitoring equipment; data transmission delay; RBF neural network.
File-URL: http://www.inderscience.com/link.php?id=116153
File-Format: text/html
File-Restriction: Access to full text is restricted to subscribers.
Handle: RePEc:ids:ijpdev:v:25:y:2021:i:2:p:175-186

Template-Type: ReDIF-Article 1.0
Author-Name: Yingji Cui
Author-X-Name-First: Yingji
Author-X-Name-Last: Cui
Title: Market product demand forecasting method based on probability statistics and convolution neural network
Abstract:
Due to the high proportion of redundant data and high prediction error rate in traditional market product demand prediction methods, this paper proposes a market product demand prediction method based on probabilistic statistics and convolutional neural network. Design the market product demand data acquisition module, collect the market product demand data, and use the improved near record sorting algorithm to clean the collected data. The statistical model of market product demand change probability is constructed to obtain the demand change probability, and the data cleaning result and demand change probability are taken as the input of the convolutional neural network model, so as to obtain the market product demand forecast result. The experimental results show that the maximum redundant data of this method accounts for only 5.9%, the prediction error rate varies between 1% and 3%, and the average prediction time is only 0.22 s. The practical application effect is good.
Journal: Int. J. of Product Development
Pages: 187-199
Issue: 2
Volume: 25
Year: 2021
Keywords: probability statistics; convolution neural network; market products; demand forecasting; sorted neighbourhood algorithm.
File-URL: http://www.inderscience.com/link.php?id=116154
File-Format: text/html
File-Restriction: Access to full text is restricted to subscribers.
Handle: RePEc:ids:ijpdev:v:25:y:2021:i:2:p:187-199

Template-Type: ReDIF-Article 1.0
Author-Name: Yanhua Wang
Author-X-Name-First: Yanhua
Author-X-Name-Last: Wang
Author-Name: Fang Zhang
Author-X-Name-First: Fang
Author-X-Name-Last: Zhang
Title: A weighted knowledge super network model for collaborative product innovation based on adjacency matrix
Abstract:
Aiming at the problems existing in traditional methods such as poor destructibility, low knowledge processing capacity and easy loss of product innovation weighted knowledge, a collaborative product innovation weighted knowledge hypernetwork model based on adjacency matrix was designed. A knowledge sharing network model is established, which is used to transform product design tasks and form multiple sub-tasks. Acquire knowledge points and carry out network modelling, acquire the membership relationship between knowledge points, design individual knowledge subgraphs, and represent different knowledge subgraphs through adjacency matrix. The hypernetwork model is constructed, and the model is used to calculate knowledge points and relevant weights of knowledge, and the knowledge fusion is realised based on the calculation results, so as to realise the construction of knowledge hypernetwork model. The results show the maximum destruction resistance coefficient of the model is about 0.8, and the maximum product innovation weighted knowledge loss rate is only 15.2%.
Journal: Int. J. of Product Development
Pages: 200-212
Issue: 2
Volume: 25
Year: 2021
Keywords: adjacency matrix; product innovation; hyper-network model; knowledge collaboration sharing; membership relationship; correlation weight.
File-URL: http://www.inderscience.com/link.php?id=116155
File-Format: text/html
File-Restriction: Access to full text is restricted to subscribers.
Handle: RePEc:ids:ijpdev:v:25:y:2021:i:2:p:200-212

Template-Type: ReDIF-Article 1.0
Author-Name: Nicholas J. Emerson
Author-X-Name-First: Nicholas J.
Author-X-Name-Last: Emerson
Author-Name: Wayne Hall
Author-X-Name-First: Wayne
Author-X-Name-Last: Hall
Title: Design for ergonomics in the bicycle industry: an engineering design concept
Abstract:
Bicycle manufacturing is an industry that has taken advantage of technological developments in materials and large-scale manufacturing for high-performance products. Higher-specification bicycles typically use fibre-reinforced composite frames that are optimised for structural performance and mass. Efforts have also been made to increase the comfort of a rider by increasing the vertical compliance of frames, but traditional fibre-reinforced composite moulding techniques restrict manufacturing flexibility in fitting frame geometry to match the requirements of an individual. Manufacturers typically subdivide their frames into sized-based grouping to reduce manufacturing, inventory and supply costs. For specific user fits, customised bicycles are produced, but only on a smaller scale. These improve comfort via improved geometric fit for the user, but typically do so at high cost, or by using sub-optimum material selections. In this paper, an innovative, segmented frame design concept is introduced to facilitate a design-for-ergonomics and mass-customisation approach in the high-volume, high-performance bicycle industry.
Journal: Int. J. of Product Development
Pages: 1-10
Issue: 1
Volume: 25
Year: 2021
Keywords: product development; engineering design; bicycle; comfort; fit; design for ergonomics; mass customisation; sports and leisure industry.
File-URL: http://www.inderscience.com/link.php?id=115699
File-Format: text/html
File-Restriction: Access to full text is restricted to subscribers.
Handle: RePEc:ids:ijpdev:v:25:y:2021:i:1:p:1-10

Template-Type: ReDIF-Article 1.0
Author-Name: Denniz Dönmez
Author-X-Name-First: Denniz
Author-X-Name-Last: Dönmez
Author-Name: Matthew J. Kerry
Author-X-Name-First: Matthew J.
Author-X-Name-Last: Kerry
Author-Name: Gudela Grote
Author-X-Name-First: Gudela
Author-X-Name-Last: Grote
Title: A meta-analysis of stability and flexibility effects on performance in product development
Abstract:
Product development teams need stability to operate efficiently and flexibility to adjust dynamically. Stability and flexibility have typically been studied in separate literatures of organisational adaptation and design, respectively. The separation of stability-flexibility has led to the two constructs as incompatibles. The current study uses the product development domain to synthesise the disparate literatures. A systematic review resulted in K = 55 analysed studies (N = 11,666). Our findings are three-fold. First, stability and flexibility exhibit a significant positive relationship. Meta-regressions support joint-positive effects with shared-variance (18%) in performance. Second, although structural forms typify stability and flexibility, stronger performance effects are observed for behavioural forms. Third, stability's stronger effect for project (vs. product) performance is evident only by complementary conceptions. That is, structural stability's project effect is strengthened by behavioural flexibility, and vice-a-versa. Conceptual conflation of stability and flexibility obscures the differential performance effects.
Journal: Int. J. of Product Development
Pages: 11-39
Issue: 1
Volume: 25
Year: 2021
Keywords: stability and flexibility; product development; performance; meta-analysis.
File-URL: http://www.inderscience.com/link.php?id=115703
File-Format: text/html
File-Restriction: Access to full text is restricted to subscribers.
Handle: RePEc:ids:ijpdev:v:25:y:2021:i:1:p:11-39

Template-Type: ReDIF-Article 1.0
Author-Name: Meng Wang
Author-X-Name-First: Meng
Author-X-Name-Last: Wang
Author-Name: Elmar Beeh
Author-X-Name-First: Elmar
Author-X-Name-Last: Beeh
Author-Name: Ping Zhou
Author-X-Name-First: Ping
Author-X-Name-Last: Zhou
Author-Name: Horst E. Friedrich
Author-X-Name-First: Horst E.
Author-X-Name-Last: Friedrich
Title: Implementing functional integration of products based on axiomatic design in the product concept phase
Abstract:
This paper proposes an approach based on the Axiomatic Design to implementing functional integration for product conception in the project early phase. At the concept stage, engineers must rely on their own experience and expertise to generate the realisable concepts and select the best one. However, the combination of products especially with complex functions is very difficult to avoid functional conflict. An approach based on the Axiomatic Design using a design matrix is proposed to formulate the functional integration process. The application process of this approach is illustrated by using exemplary three by three matrices. This interpreted design principle may work as the fundamental for the realisation of product design based on design matrices. In the case study, the proposed approach is applied in suspension conception of electric cars which combines electric motors into the longitudinal arms of twist beam axles.
Journal: Int. J. of Product Development
Pages: 40-61
Issue: 1
Volume: 25
Year: 2021
Keywords: design method; axiomatic design; functional integration; suspension design.
File-URL: http://www.inderscience.com/link.php?id=115706
File-Format: text/html
File-Restriction: Access to full text is restricted to subscribers.
Handle: RePEc:ids:ijpdev:v:25:y:2021:i:1:p:40-61

Template-Type: ReDIF-Article 1.0
Author-Name: Dag Bergsjö
Author-X-Name-First: Dag
Author-X-Name-Last: Bergsjö
Author-Name: Amer Catic
Author-X-Name-First: Amer
Author-X-Name-Last: Catic
Author-Name: Daniel Stenholm
Author-X-Name-First: Daniel
Author-X-Name-Last: Stenholm
Title: A lean framework for reusing knowledge: introducing engineering checksheets
Abstract:
In this study, we have brought several knowledge management theories and practices together and have investigated the valuable aspects in each one of them. Throughout our research, we have enabled knowledge reuse to be a guiding principle and have attempted to formulate an improved method for documenting knowledge. The focus has been to create actionable and reusable knowledge presented at the right time, to the right audience in a digitally condensed format that may hopefully will lead to improved decision making, thereby potentially driving innovation and effectively reducing overall product realisation lead-time. We have explored the effect of the engineering checksheet concept in two separate case studies where it was implemented in a real-life setting. The concept showed positive results as a knowledge carrier both for reuse by experienced and novice users.
Journal: Int. J. of Product Development
Pages: 283-302
Issue: 3
Volume: 25
Year: 2021
Keywords: organisational learning; knowledge management practice; knowledge sharing; explicit knowledge; decision support; case studies; engineering checksheet; ECS.
File-URL: http://www.inderscience.com/link.php?id=118010
File-Format: text/html
File-Restriction: Access to full text is restricted to subscribers.
Handle: RePEc:ids:ijpdev:v:25:y:2021:i:3:p:283-302

Template-Type: ReDIF-Article 1.0
Author-Name: Alessandra Cantini
Author-X-Name-First: Alessandra
Author-X-Name-Last: Cantini
Author-Name: Filippo De Carlo
Author-X-Name-First: Filippo De
Author-X-Name-Last: Carlo
Author-Name: Mario Tucci
Author-X-Name-First: Mario
Author-X-Name-Last: Tucci
Title: Serious game for lean layout planning: a proposal for involving company staff within the design process
Abstract:
Planning the layout of a company directly affects the performance of a production system. To generate an effective layout solution, it is fundamental to involve both design experts and company staff at all stages of the design process. This paper presents a novel lean layout planning system to engage company staff in the research of the layout solution. A two-step process was followed: first, a literature-based approach was applied to get a prototype layout, then a strategy to engage the company's staff in the achievement of the final solution was proposed. The article describes the proposed strategy and shows how the layout changes with or without the involvement of skilled staff. A case study of a leather bags manufacturing plant is also presented. The case study results highlight how the proposed procedure can be helpful in achieving a solution that better satisfies workers and makes employees accept working changes.
Journal: Int. J. of Product Development
Pages: 62-84
Issue: 1
Volume: 25
Year: 2021
Keywords: lean layout planning; serious game; facility layout problems; lean manufacturing; lean design; leather bags manufacturing plant.
File-URL: http://www.inderscience.com/link.php?id=115707
File-Format: text/html
File-Restriction: Access to full text is restricted to subscribers.
Handle: RePEc:ids:ijpdev:v:25:y:2021:i:1:p:62-84