Elsevier

Information Sciences

Volume 373, 10 December 2016, Pages 557-570
Information Sciences

Monitoring and assessing fruit freshness in IOT-based e-commerce delivery using scenario analysis and interval number approaches

https://doi.org/10.1016/j.ins.2016.07.014Get rights and content

Abstract

We are concerned with the monitoring and assessment of in-transit fruit freshness in e-commerce deliveries. After comparing the fulfillment processes of fresh fruit transportation in traditional retailing and e-retailing, we formulate an Internet of Things–based framework for monitoring fruit e-commerce deliveries. Based on the fulfillment operations and monitoring modules of the framework, we propose an approach based on a two-stage scenario for assessing the freshness of in-transit fruits. In the first stage, we use a learning-by-doing mechanism to develop a scenario construction method to automatically obtain the most appropriate delivery environment and the occurrence probability for each scenario. In the second stage, we integrate the interval comparison technique into the scenario analysis method to address the freshness assessment of in-transit fruits. The effectiveness and advantages of our approach are verified using numerical simulations.

Introduction

With the development of e-commerce and advanced information technologies, retailing has experienced a revolution. Currently, e-retailing has advantages over traditional retailing in many areas, such as books, music, clothing and electrical appliances. However, fresh produce presents many challenges for e-retailing due to characteristics such as perishability and high-cost logistics, especially in developing countries [3], [17]. Recently, the Chinese government and e-commerce giants have tried to break into fresh produce e-retailing. One of the difficulties encountered is determining how to monitor and control the freshness of in-transit fresh produce [12]. Obviously, these tasks are especially difficult without help from advanced information technologies [16], [20], [30]. In an empirical study, Shin and Eksioglu [32] observed that the application of radio frequency identification devices (RFIDs) considerably improves labor productivity in U.S. retail supply chains.

In the literature, some recent studies have applied related information technologies to the monitoring and control of in-transit fresh produce, presenting specific solutions. For example, Ruiz-Garcia et al. [28] applied ZigBee-based wireless sensors to the monitoring of fruit logistics, analyzed the battery life of the sensors, and evaluated the reliability of the whole system. Abad et al. [1] presented a monitoring system consisting of a smart RFID tag subsystem located in the delivery trucks and a commercial reader/writer subsystem located in the delivery nodes, reporting advantages of this system in an intercontinental fresh fish supply chain. Kang et al. [11] presented a simulation method to assess the performance of optimization models and to determine the key parameter values in a RFID sensor–based cold transportation system, while Mainetti et al. [18] applied radio frequency technologies and EPCglobal standards to develop a traceable system of fresh vegetable products. Mejjaoulia and Babiceanu [19] presented an integrated RFID sensor network system for optimizing the transportation of perishable products, indicating that the system was helpful for reducing operation costs. Xiao et al. [40] argued that heavy sensory data needs in traditional monitoring systems reduced data transmission efficiency; thus, they integrated a wireless sensor network (WSN) with compressed sending (CS) to develop a temperature monitoring system for frozen and chilled aquatic products. Finally, Trebar et al. [37] used RFID sensors to monitor the temperatures of styrofoam boxes during the transport of fresh fish and proposed a time- and energy-saving method for packing fish.

The above studies have made specific contributions to the monitoring and control of fresh produce by applying advanced information technologies and providing practical support for the transportation of fresh produce. However, most studies have focused on conventional supply chains, paying little attention to the e-commerce delivery of fresh produce. In e-retailing, the delivery of fresh produce is quite different from traditional transportation, which creates new challenges [13]. Meanwhile, different kinds of fresh produce have specific characteristics in e-fulfillment [16]. Motivated by these observations, in this work, we are concerned with the monitoring of fresh fruit in e-commerce deliveries.

Monitoring and control systems aim to maintain the freshness of perishable products while in transit. However, loss of freshness may result at any point during the e-fulfillment process of fresh fruit orders. Thus, real-time assessment of these products is key to controlling their freshness. The e-fulfillment of fruit orders often involves multiple participants conducting their respective delivery services [25]. The complexity of and uncertainty in this process make monitoring and assessing of the freshness of in-transit fruits difficult. Scenario analysis is an effective method for analyzing complex uncertainty by considering both possible events and their occurrence probabilities [5], [8], [15]. As mentioned above, the e-fulfillment of fresh fruit orders involves multiple operation links and service participants, creating a process that is full of uncertainty. The freshness of in-transit fruits will vary in different fulfillment situations. Thus, scenario analysis is suitable for recognizing different e-fulfillment situations for fruit orders.

Similarly, it is increasingly popular to develop models using fuzzy or interval information in the literature due to uncertainty [14], [31], [34], [35]. The data, such as temperature and humidity information, produced by fruit monitoring sensors are often interval values. Classic scenario analyses cannot be directly used to address these interval data in each possible scenario. Fortunately, a body of literature addresses interval comparison methods [4], [45], [47]. The integration of interval comparison methods into classic scenario analysis may provide feasible solutions for assessing the freshness of in-transit fruits in different interval situations.

Motivated by the above observations, in this work, we present an Internet of Things (IOT)-based framework for monitoring fruit e-commerce delivery. We use scenario analysis methods to construct delivery scenarios in the e-fulfillment of fruit orders to provide corresponding assessments of freshness at each step in the delivery process. Meanwhile, based on the characteristics of the sensor data, we integrate interval comparison into the scenario-based freshness assessment approach.

To sum up, in this work, we make the following contributions:

  • (i)

    We compare fresh fruit transportation in traditional retailing and e-retailing and present an IOT-based framework for monitoring fruit e-commerce deliveries. Detailed descriptions of fruit e-fulfillment operations, as well as of the modules in the IOT-based monitoring framework, provide a basis for developing a scenario-based approach to assessing in-transit fruit freshness.

  • (ii)

    We divide the freshness assessment scenario into a scenario construction stage and a freshness assessment stage. We then use a learning-by-doing mechanism to develop a scenario construction method that can automatically obtain both the most suitable environment for each scenario and the occurrence probability of each scenario from practical delivery operations.

  • (iii)

    We integrate interval comparison techniques into scenario analysis methods to assess the freshness of in-transit fruits. This integrated approach assesses freshness by comparing the similarity of the sensing environment to the most suitable environment for each scenario.

The remainder of this paper is organized as follows. Section 2 presents an IOT-based framework for monitoring fruit e-commerce deliveries, providing the basis for the scenario analysis portion of the freshness assessment. In Section 3, we propose a scenario-based approach for assessing the freshness of in-transit fruits, which is divided into a scenario construction stage and a freshness assessment stage. Section 4 presents the experimental results showing the effectiveness and advantages of our contributions. Section 5 concludes the work and provides recommendations for future study.

Section snippets

Fruit e-commerce delivery

In traditional commerce, fresh fruits are often sold to end consumers through multiple intermediaries, such as wholesalers, distributors and retailers, as Fig. 1 shows. In the traditional retailing channel, fresh fruits are first picked on farms and then transported to local processing centers (LPCs) where fruits are washed, cooled, packaged, and so on. Then, the processed fruits are often purchased by wholesalers who will either store the fruits in their warehouses or directly transport them

A scenario-based approach for assessing the freshness of in-transit fruits

As mentioned in the Introduction, scenario analysis is a prediction and assessment technique for analyzing both the possible events and the occurrence probability of each event. This technique has been widely used for complicated decision-making issues. Although fresh fruit e-retailing is significantly simpler than traditional fruit retailing, uncertainties remain, and loss of freshness will differ based on the operations and environments of the e-fulfillment process. Thus, in this work, we

Numerical simulations

In this section, we use numerical experiments to demonstrate the efficiency and advantages of applying our two-stage approach to assessing the freshness of in-transit fruits in an IOT-based framework.

Concluding remarks

In this work, we investigated the monitoring and assessment of fruit freshness in e-fulfillment. For the monitoring aspect, we provided a comparison of the traditional fruit retailing and e-retailing transportation processes and applied IOT-related technologies, such as GPS, RFID sensor tagging and wireless sensor networks, to formulate a monitoring framework for e-commerce fruit delivery. For the fruit freshness assessment, we presented a two-stage scenario-based approach for assessing the

Acknowledgments

This research is supported by the National Natural Science Foundation of China (Nos. 71471025 and 71531002), the Hong Kong Scholars Program Mainland-Hong Kong Joint Postdoctoral Fellows Program (No. G-YZ87), the Doctoral Scientific Research Foundation of Northwest A&F University (No. 2452015325), the Natural Science Basic Research Project in Shaanxi Province (No. 2016JQ7005) and China Ministry of Education Social Sciences and Humanities Research Youth Fund Project titled as “Delivery

References (47)

  • LiuX. et al.

    Perceptions of information gaps in farm-to-table studies

    Food Cont.

    (2015)
  • M. Llarena-Reino et al.

    Horizon scanning for management of emerging parasitic infections in fishery products

    Food Cont.

    (2015)
  • L. Mainetti et al.

    An innovative and low-cost gapless traceability system of fresh vegetable products using RF technologies and EPCglobal standard

    Comput. Electron. Agric.

    (2013)
  • R.R. Oliveira et al.

    An intelligent model for logistics management based on geofencing algorithms and RFID technology

    Expert Syst. Appl.

    (2015)
  • Y. Ouyang et al.

    A new model for intuitionistic fuzzy multi-attributes decision making

    Eur. J. Oper. Res.

    (2016)
  • W. Pedrycz

    Why triangular membership functions?

    Fuzzy Sets Syst.

    (1994)
  • W. Pedrycz

    From fuzzy data analysis and fuzzy regression to granular fuzzy data analysis

    Fuzzy Sets Syst.

    (2015)
  • P.J. Reaidy et al.

    Bottom-up approach based on internet of things for order fulfillment in a collaborative warehousing environment

    Int. J. Prod. Econ.

    (2015)
  • RuanJ. et al.

    Relief supplies allocation and optimization by interval and fuzzy number approaches

    Inf. Sci.

    (2015)
  • L. Ruiz-Garcia et al.

    Performance of zigbee-based wireless sensor nodes for real-time monitoring of fruit logistics

    J. Food Eng.

    (2008)
  • A. Sengupta et al.

    On comparing interval numbers

    Eur. J. Oper. Res.

    (2000)
  • S. Shin et al.

    An empirical study of RFID productivity in the u.s. retail supply chain

    Int. J. Prod. Econ.

    (2015)
  • M. Srbinovska et al.

    Environmental parameters monitoring in precision agriculture using wireless sensor networks

    J. Clean Prod.

    (2015)
  • Cited by (0)

    View full text