Risk management in perishable food distribution operations. A distribution route selection model and whale optimization algorithm
| Pages | 291-311 |
| Date | 05 October 2019 |
| Published date | 05 October 2019 |
| DOI | https://doi.org/10.1108/IMDS-03-2019-0149 |
| Author | Yingchao Wang,Chen Yang,Hanpo Hou |
| Subject Matter | Information & knowledge management |
Risk management in perishable
food distribution operations
A distribution route selection model and
whale optimization algorithm
Yingchao Wang
School of Business, Beijing Technology and Business University, Beijing, China
Chen Yang
Beijing Institute of Technology, Beijing, China, and
Hanpo Hou
School of Business, Beijing Technology and Business University, Beijing, China
Abstract
Purpose –The purpose of this paper is to predict or even control the food safety risks during the distribution
of perishable foods. Considering the food safety risks, the distribution route of perishable foods is reasonably
arranged to further improve the efficiency of cold chain distribution and reduce distribution costs.
Design/methodology/approach –This paper uses the microbial growth model to identify a food safety
risk coefficient to describe the characteristics of food safety risks that increase over time. On this basis, with
the goal of minimizing distribution costs, the authors establish a vehicle routing problem with a food safety
Risk coefficient and a Time Window (VRPRTW) for perishable foods. Then, the Weight-Parameter Whale
Optimization Algorithm (WPWOA) which introduces inertia weight and dynamic parameter into the native
whale optimization algorithm is designed for solving this model. Moreover, benchmark functions and
numerical simulation are used to test the performance of the WPWOA.
Findings –Based on numerical simulation, the authors obtained the distribution path of perishable foods
under the restriction of food safety risks. Moreover, the WPWOA can significantly outperform other
algorithms on most of the benchmark functions, and it is faster and more robust than the native WOA and
avoids premature convergence.
Originality/value –This study indicates that the established model and the algorithm are effective to
control the risk of perishable food in distribution process. Besides, it extends the existing literature and can
provide a theoretical basis and practical guidance for the vehicle routing problem of perishable foods.
Keywords Risk management, Supply chain management, Vehicle routing problem,
Whale optimization algorithm
Paper type Research paper
1. Introduction
With the rapid development of information technology, supply chains for fresh agricultural
foods under the Online To Offline model are increasing. From Freshhema to JD, fresh food
retailing is becoming the next frontier for China’s e-commerce giants, and offline physical
retail and online e-commerce enterprises are actively exploring new retail models. Between
2012 and 2018, the value of China’s fresh food e-tail market grew from under RMB4bn
($580m) to almost RMB105bn ($15.7bn), according to an analysis from the China
International Electronic Commerce Center (2018). However, the prosperity of fresh food
e-commerce also brings great challenges to cold chain distribution operations, demonstrated
by extended coverage in newspapers and transport journals: food safety, food quality and
sustainability. It is well known that the distribution of perishable food is different from the Industrial Management & Data
Systems
Vol. 120 No. 2, 2020
pp. 291-311
© Emerald PublishingLimited
0263-5577
DOI 10.1108/IMDS-03-2019-0149
Received 17 March 2019
Revised 20 May 2019
12 August 2019
Accepted 28 August 2019
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/0263-5577.htm
The authors would like to thank the strong support provided by the Beijing Social Science Foundation
Project (No. 17GLB015), Beijing Institute of Technology Research Fund Program for Young Scholars
and the Graduate Research Capacity Improvement Program in 2019.
291
Perishable
food
distribution
operations
distribution of other products. Food products exhibit consistent quality changes throughout
the supply chain, up to final consumption. As people’s health awareness continues to
increase and the constant pursuit of high-quality life continues, customers’requirements for
food freshness are more stringent than ever. Therefore, quality, health and safety require
central consideration in perishable food distribution to control food safety risks.
Risk managementhas been applied in various fields, suchas marketing (Rink et al., 2013),
finance (Clarke et al., 2017) and supply chains (Ping et al.,2017).Combinedwithmany
descriptions of risk, this paper defines risk as “Something happened as a result of not
accurately knowing future events.”For food supply chain operations, risks come from the
characteristics of the food itself; assuringfood quality and safety has becomea major subject
in food supplychain management and a majorfocus for authorities andprofessionals (Ye et al.,
2019). The quality of food involves certain characteristics, such as the appearance, texture,
taste, smelland nutritional value of the item,which must be ensured to satisfyconsumers. For
example, T.T.T. theory (Deeds, 1959) advocates that the quality of fresh and perishable foods
depends on the circulation time, storage temperature and tolerance of the food itself. In
addition, the food quality depends to a greater extent on the concentration of diseased
microorganisms, e.g., the presence of salmonella in chickenor cows infected with BSE has led
to seriousillnesses, even death and majorproduct recalls (Akkerman et al., 2010). Notably, it is
impossible to completely eliminate food contaminants (Marvin et al., 2009). Therefore, it is
necessary to optimize the whole operation of the food supply chain, including theproduction
operations, the storage operations, the distribution operations, circulation and processing, to
ensure food safety and quality as much as possible (Diabat et al., 2012).
Due to the aforementioned facts, an innovative distribution delivery route that carefully
considers foodsafety risk must be made on the basis ofthe vehicle routing problem (VRP),so
that customers can obtain the freshest food products and logistic companies can have a low
delivery cost. The VRP is a classic operational research problem raised in 1959 (Dantzig and
Ramser, 1959), which mainly considers the lowest comprehensive cost in the distribution
process. In 1987, the VRP with time windows (Solomon, 1987) was first studied to further
expand the scope of the VRP. In theory, problems such as the VRP can be solved by a
mathematicalprogramming solution in operationsresearch, but the actual solution processis
usually very complicated. Fortunately, with the development of computer technology,
meta-heuristic algorithms such as particle swarm optimization (PSO) (Li et al., 2019), the
genetic algorithm(Assaf and Saleh, 2017) and the simulated annealing algorithm (Azadet al.,
2017) are graduallybeing applied to solve the VRP in a very shortamount of time. Meanwhile,
the whale optimization algorithm (WOA) (Mirjalili and Lewis, 2016) is a new kind of
meta-heuristic algorithm with the advantages of simple algorithm principles and few control
parametersand has been widely used in many fields (Mafarjaand Mirjalili, 2017). It providesa
new way to solve the distribution route selection model that considers food risks.
Therefore, the main contributions of our research are as follows:
(1) To measure the food safety risks in the distribution process, this paper focuses on
the impact of microorganisms on food safety and establishes a food safety risk
coefficient based on the microbial growth model to characterize the increasing food
safety risks over time. Moreover, safety thresholds are used to control food safety
risks in distribution operations. Based on this concept, a Vehicle Routing Problem
with a food safety Risk coefficient and Time Window (VRPRTW) was established
for perishable food, which mainly considers a single distribution center that
distributes single goods to multiple demand points and assumes that the demands of
customers are known in advance.
(2) The Weight-Parameter Whale Optimization Algorithm (WPWOA) which introduces
inertia weight and dynamic parameter into the native WOA is used to solve
292
IMDS
120,2
To continue reading
Request your trial