Smart Warehouse Management System Concept With Implementation

Download as pdf or txt
Download as pdf or txt
You are on page 1of 6

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/329901029

Smart Warehouse Management System Concept with Implementation

Conference Paper · November 2018


DOI: 10.1109/NEUREL.2018.8587004

CITATIONS READS

2 905

5 authors, including:

Emir Zunic Sead Delalić


Info Studio d.o.o.¸Sarajevo University of Sarajevo
34 PUBLICATIONS   62 CITATIONS    7 PUBLICATIONS   8 CITATIONS   

SEE PROFILE SEE PROFILE

Kerim Hodžić Admir Beširević


University of Sarajevo University of Sarajevo
9 PUBLICATIONS   17 CITATIONS    7 PUBLICATIONS   17 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Warehouse Process Optimization in Bosnia and Herzegovina View project

All content following this page was uploaded by Kerim Hodžić on 25 December 2018.

The user has requested enhancement of the downloaded file.


Smart Warehouse Management System Concept
with Implementation
Emir Žunić, Sead Delalić, Kerim Hodžić, Admir Beširević, Harun Hindija

 analysed and optimized separately, but it is also important


Abstract — Distribution companies use complex software to analyse and optimize them as connected parts of the
systems called WMS (Warehouse Management System). The whole process.
WMS is an important part of the company’s business and it
can make processes simple to keep track of. Smart WMS
optimizes processes to save resources and to create a more
efficient working place. This paper describes the concept of
a smart WMS that is implemented in one of the largest
distribution companies in Bosnia and Herzegovina. The
system uses artificial intelligence and optimization
algorithms to improve working process. The paper describes
the complete warehouse workflow that includes stock
planning, initial product placement, transfer from stock to
pick zone, order picking process, transport and tracking.
The anomaly detection is used in some processes to improve
the whole system. The main contribution of this paper is the
presentation of an efficient and in the real world used smart
WMS concept. Fig. 1. Main WMS processes
Keywords — Smart Warehouse Management System,
Stock Planning, Order Picking Process, Location When the goods arrive, they need to be initially placed
Management, Vehicle Routing Problem
in the warehouse in a proper position. Warehouses can

THE b
I. INTRODUCTION have many different layouts. Warehouse layout mostly
depends on the building shape. Often, the complete layout
asic purpose of the warehouse is to store is divided into sectors of the same or different shape.
products in one place and transfer them forward to Sectors are created to separate different brands or to
company customers. Depending on the type of the create different temperature conditions. Each sector
company, those products can be ordered and shipped from contains shelves that can be arranged in different shapes.
manufacturers or manufactured in-house. The complete Warehouses can have separate entrances and exits. The
distribution process consists of a number of smaller ones entrance is a part of the warehouse where the goods are
and they must be properly monitored. Therefore, the brought from the manufacturers and the exit location is
software solutions called Warehouse Management the part of the warehouse from where the orders are
Systems (WMS) are created. If the Warehouse transferred to the final customer. Shelves are organized by
Management System is implemented well, it can save arranging many pallet places in a line and also in levels to
large amounts of resources to distribution/manufacturing enlarge warehouse capacity. Shelves are divided into
companies by improving the complete working process stock and picking zone. Picking zone, sometimes referred
and providing useful business analysis. Many large to as the golden zone, is part of the warehouse racks from
manufacturing companies require that their distributors where products can be handpicked. Higher racks are
have implemented WMS. Complete product distribution called the stock zone and products from the stock zone are
process is usually consisted of five chronologically usually picked using forklifts.
connected steps shown in Fig. 1 and each step can be It is important to plan which and how much of every
product to store in the warehouse. Storing quantities of
Emir Žunić is with the Info Studio d.o.o. Sarajevo and Faculty of products that are larger than the quantity that a company
Electrical Engineering, University of Sarajevo (e-mail:
[email protected]). can sell can cause unnecessary costs in terms of
Sead Delalić is with the Faculty of Science, University of Sarajevo warehouse space, warehouse worker paying hours and
and Info Studio d.o.o. Sarajevo (e mail: [email protected]). electricity costs. On the other hand, not storing enough
Kerim Hodžić is with the Faculty of Electrical Engineering,
University of Sarajevo and Info Studio d.o.o. Sarajevo (e-mail: quantities of product can lead to the company not being
[email protected]). able to deliver ordered quantities, which can lead to profit
Admir Beširević is with the Faculty of Science, University of and customer loss.
Sarajevo and Info Studio d.o.o. Sarajevo (e-mail:
[email protected]). Customer orders are coming during the day and
Harun Hindija is with the Faculty of Science, University of Sarajevo warehouse workers need to collect ordered products. The
and Info Studio d.o.o. Sarajevo (e-mail: [email protected]). workers are going through the aisles between shelves to

978-1-5386-6974-7/18/$31.00 ©2018 IEEE


pick the ordered items. The described process is called the tracking. In [3], it is described how the RFID can improve
order picking. The workers usually use the picking cart to the efficiency of the warehouse management and make a
make the process faster and easier and to be able to travel rapid self-recording of receiving and delivery.
between aisles and cross-aisles. Orders usually contain The objective of paper [4] was to propose and analyse
multi-brand items that are placed in different sectors. results of an Internet of things (IoT)-based warehouse
Products stored on the top levels not available in the management system with analytical approach using
picking zone are not reachable for workers to take them computational intelligence techniques to enable
by hand and they have to wait for a forklift to come to warehouse smart logistics. Because of the wide range of
transfer goods from the stock to the picking zone. That approaches and algorithms that can be used in WMS,
consumes an additional time for order to be picked.
evaluating their effectiveness is an important aspect to be
When all ordered items are collected, they have to be
covered. In paper [5], a simulation model is proposed to
transferred to customers, usually by transporting vehicles.
evaluate the performance of a warehouse system for
A set of transporting vehicles is usually called the
transport fleet. Orders are grouped by customers, usually perishable products taking into account the effect of the
by their location and distance. Each group is assigned to a uncertainties which typically affect the supply chain.
vehicle in the fleet.
III. CASE STUDY
Each described process contains the set of smaller
processes that can be optimized. This paper describes a In this section, every cardinal warehouse operation is
concept of the smart warehouse management system described and a concrete solution for the operation
implemented in one of the largest distribution companies improvement is proposed. To implement smart warehouse
in Bosnia and Herzegovina. The concept includes many management system, prerequisite is to have developed a
algorithms for optimizing the complete warehouse warehouse management system that includes a database
management system. with history of transactions and product locations. Then,
This paper consists of five sections. In the first section, collected information are used for constructing algorithms
an introduction is given and the complete warehouse and solutions around it. In this particular case, before the
working concept is described. In the second section, a implementation of the smart WMS was conducted,
literature review of related work is given. In the case information system for distributing company was
study section, proposed concept is described. Then, the implemented.
results are given and described for optimized processes. Every rack was labelled by barcode which represents a
In the final section, the conclusion and future work ideas location unique identifier. Some important functionalities
are presented. that information system provided was storing history
about every internal and external order, transaction and
II. RELATED WORK product location change. As a part of the system, the
Every aspect of the warehouse management system can Android application whose main goal was to provide
be analysed and optimized separately. That is conducted warehouse workers easier way to register important
in Section III. In this section, a few publications that are information to the system in a way that does not affect
analysing warehouse management systems in general are their work was developed. Application screens for
mentioned and briefly described. registering new product and product location change are
First step for improvement of the traditional warehouse shown in Fig 2.
management system can be achieved by automating
processes. Article [1] investigates the impact of a
warehouse management system on supply chain
performance. The supply chain procedures carried out in
the warehouse were reviewed before and after
implementing software that handles warehouse
transactions.
More improvement can be achieved by implementing
more advanced data mining and artificial intelligence
techniques in warehouse processes. Paper [2] proposes
similar alternative called intelligent WMS (i-WMS). That
system consists of five subsystems: intelligent logistic
system, intelligent warehouse system, real-time
transportation monitoring, sales forecasting system, and
Fig. 2. Android application screens
intelligent sales summary system. Research conducted in
that paper integrates state-of-the-art results in the field of
intelligent systems-neural network, bee colony A. Stock planning
optimization, fuzzy control, and decision support system - For every distributing company, key for success is to
together with the Radio Frequency Identification (RFID) provide more quality service to every customer while
technology. Many similar aspects are also covered in this spending fewer resources. In order to serve every
paper, analysed from the different perspective and with customer order, there has to be enough units of every
different algorithms and methods implemented. ordered product in stocks. The solution is not just to store
Barcodes or RFIDs are often used in WMS for product
more and more product units, because space and human product units to be picked by the warehouse workers.
resources are limited. Usually, in the mid-sized warehouses, the worker is
Stock has to be planned accordingly for the future. This walking through the warehouse and collecting product
is an old problem in economy known as demand units in a picking cart. Orders can contain dozens of
forecasting and many different algorithms and approaches different products so determining the best path for
are used for solving it. Study explained in [6] attempts to collecting order is not an easy task for a human,
identify the main determinants of forecasting accuracy especially if the important information like product unit
and gives directions which method to use on what kind of expiry date is considered. So, as part of the smart
data through simulations and empirical investigations warehouse management system, a module that gives
involving 14 popular forecasting methods. workers a sorted list of products and their locations that
An important aspect in demand forecasting is product he needs to collect in a suggested order with the earliest
seasonality. Different products are showing different expiration dates first is implemented.
seasonal patterns. Another important aspect that must be As it is described in a paper [8], the order picking
considered are discounts. In solution described in this problem can be easily transformed to the traveling
paper, the module for demand forecasting is integrated salesman problem (TSP). It can be stated that the TSP is
and it uses X-13 seasonality filter and anomaly detection solved for practical cases by using the 2-opt and 3-opt
algorithm for detecting and filtering values that do not fit algorithms. Those algorithms require the ability to
into the pattern. calculate the distance between two items. The algorithm
Algorithm is used to forecast monthly demand for every that uses the dynamic programming is used to calculate
product based on twenty years history of data and can also the shortest distances between two pallet places where the
cope with discounts if they are announced at least one items are placed. The generalization of the algorithm for
month in advance. distance calculating is given in [10]. This distance
calculating generalization enabled the generalization of
B. Receiving/Location Management
the algorithm for order picking. The implemented
When products arrive to the warehouse, they need to be algorithm can make an optimal path for a warehouse of an
placed in appropriate location. The effective product arbitrary and non-standard design.
positioning can improve the order picking process by The described concept requires the worker to scan a
shortening walking routes for workers, which gives them picked article, so the WMS can update the stock data and
more time to handle other warehouse operations. Solution order data.
for this problem can differ for an empty and already
operative warehouses not being able to stop their work for D. Shipping/Transport
some period of time and reorganise product locations. The Vehicle Routing Problem (VRP) and optimal
As a part of the smart WMS, location management utilization of the transport fleet has been well explored
module is implemented as set of SQL procedures that and constantly improved in the last few decades.
suggest a list of sorted positions to which every product The VRP is researched in the literature dealing with the
should be placed when it comes into the warehouse. The operational research. The problem could be described
main goal of the algorithm is to provide convergence of briefly in this way: for a given set of customers, for a
more frequent products to locations that are closer to known amount of goods to be delivered to each customer
location where prepared orders are stored. In most cases it at a precisely determined time interval, and a known fleet
is the exit location. The algorithm uses historical data for of available vehicles together with their constraints, an
calculation. More details about the algorithm optimal set of routes should be determined, where each
implementation, results and how the distances are route starts from the warehouse, each customer is served
calculated can be found in papers [7] and [8]. Another with exactly one vehicle, all customers are served, and the
important aspect in product location management is given set of routes has the smallest resource cost of all
picking zone planning. If there are enough product units possible route combinations. It should be emphasized that
in the picking zone, then workers collecting orders do not the set of routes can be optimized by different criteria
have to use forklifts to get items from higher rack floors (minimal fuel consumption, minimal number of vehicles,
and that leads to less fuel and time consumption. minimal total travel route of all vehicles and similar). It
For this purpose, the picking zone prediction algorithm has been shown that such a problem belongs to a group of
is implemented and it, by using the historical transaction so-called NP-hard problems, which are known to be
data, predicts which products and how many product units difficult to solve with very powerful computer systems,
are going to be ordered tomorrow, so the workers can and their manual solution without the use of information
prepare the picking zone a day before. The algorithm is technology is not a realistic option.
implemented as a set of SQL procedures and is based on The transport optimization and tracking is very
shifted (0.9) median value of historical data describing important part of this smart WMS implementation and it
orders and adaptive coefficients. More details about the is used to improve the final warehouse operation, which is
algorithm implementation in six variants and comparative delivering goods to final customers. The implemented
results can be found in paper [9]. algorithm is consisted of three phases. The transformation
of the input data as the time window and distance of the
C. Order picking
customers is made, which allows the unloading time for
Customers are making orders on a daily basis. each customer to become zero, which simplifies the
Information about orders are stored by the warehouse problem. In the second phase, an initial solution is
management system as a list of products and numbers of created. Then, this solution is attempted to improve
iteratively using the Tabu Search. A VRP module is Anomaly detection conducted for most exported
created as a part of the web application where users can product sales is shown in Fig. 4.
input parameters and analyse suggested transport routes.
The algorithm has been fully implemented in a way
that supports a very large number of realistic constraints.
By doing so, the realized algorithm is adapted to practical
applications and problems, including practical limitations
that the real situation undoubtedly contains. The
aforementioned system provides the ability to maximize
the cost of transport, optimize the fleet of used vehicles,
and, based on simulated situations, obtain the best
solution for future vehicle acquisitions.
One part of implemented smart WMS is the GPS
tracking module. The module is used to track vehicles by Fig. 4. Anomaly detection module
using the data sent by the GPS tracking device. Data
gathering application, which communicates with devices, IV. RESULTS
parses the received data and saves the data do the The concept described is implemented in one of the
database. It is developed in the Java programming largest distribution companies in Bosnia and
language, by using multithreading to allow concurrent Herzegovina. The first improvement for company
execution for maximum utilization and flexibility. business was made by implementing the warehouse
Module for GPS data analysis is developed as part of management system with a centralized database that
the integrated web application and it is shown in Fig. 3. stores information like product location, orders and
transaction history etc. Although, there is no metric to
measure it, improvement in everyday work was
noticeable. The next part was the process of implementing
artificial intelligence algorithms to every step of the
complete process from the very beginning, which is the
process of order planning and product positioning, to the
end, which is the transporting products to the customers.
Order picking optimization module described in [8]
was implemented first and it showed the improvement of
more than 40% in average route length. After initial
success, the company decided to start using the algorithm
Fig. 3. GPS tracking application screen in other warehouses and some of them had a non-standard
layout. The algorithm was improved and adapted to work
Analysing driven routes and comparing them to the on different warehouse layout and that process is
suggested routes gives information about which described in [10]. It was noticed that there is more place
customers were served, information about the traveling for improvement if product locations were organised
time between customers and loading time per order and better, so the module for optimal product positioning
product. This information is gathered through time and described in [7] was implemented. One of the most
used to improve parameter values used in the significant cost savings is that the number of items with
implemented vehicle routing algorithm. the expiration date is reduced to a minimum.
E. Other improvements In Table 1, the analysis conducted one and nine months
after the algorithm is implemented are shown. It can be
The warehouse management system contains a large set
noticed that average order picking route length reduced
of data that are the result of the user input: orders data,
most for smaller orders through time. Reason for that is
number of articles, articles description (mass, volume,
more space for improvement in orders containing less
dimensions) etc. Sales prediction is based on the analysis
products and also slow product movements to their
and usage of the previous sales data. Consequently, the
suggested location only when positions become free to
data accuracy is crucial for the correct execution of the
occupy.
process and control of the stock. The proposed smart
WMS concept uses the anomaly detection algorithms for TABLE I: AVERAGE PICKING ROUTE LENGTH REDUCTION
analysis and corrections of the order data. Number of Average length Average length
The type of order data is time-series. Adaptation of a ordered items reduction after one reduction after
Twitter algorithm for detecting anomalies described in month (%) nine months (%)
[11] is used in the proposed smart WMS concept. The 30-40 17.34 17.3
20-29 12.5 11.4
Twitter algorithm reveals seasonality and trend in data,
10-19 10 10.2
which makes it suitable for use in the product ordering 5-9 7.1 12.6
systems. Early detection of anomalies can reduce the 0-4 4.4 14.5
number the cancelled orders. Detected anomalies that are
caused by mistake are eliminated from records used in the The next important aspect to improve was determining
sales predictions and in the stock planning module in which products and how much of them to put in the
order to improve the prediction results. picking zone so they can be easily accessed. Algorithm
View publication stats

for the picking zone capacity and content prediction using warehouse management system by giving optimized
order history described in [9] was implemented. It solutions to users. The implemented concept optimizes
provided more than 90% ordered product amounts to be stock planning, initial product placement, stock to picking
picked from picking zone and also reducing transfers zone transfer process, as well as order picking, transport
from stock to pick zone by 16.2 times and also reduced and tracking processes.
the total number of transactions by more than 4%. The future work plan is to implement optimization
Anomaly detection is used to detect irregularities like techniques in other processes important for the
input or calculation mistakes in information about the distribution business. The portal where customers can
customer orders. Currently, every anomaly detected by make orders and get recommendations based on their
the algorithm is cross-checked by the company personnel. preferences will be implemented. The anomaly detection
From the total number of orders, approximately 3.55% algorithm will be improved and adjusted for the usage in
were detected by the algorithm as anomalies. From that other warehouse processes.
number, 67% were detected as correct by the company The VRP is going to be improved by adding more
personnel. After this phase and algorithm adjustment it is constraints to implemented optimization techniques. The
planned to incorporate it in other parts of the WMS that GPS tracking app will be improved with additional
contain user inputs or calculations. analyses and algorithms for smart tracking and route
The stock planning algorithm is currently in a testing comparisons. The algorithm for order picking will be
phase. Every 2 weeks when orders are made, the improved by adding additional constraints as weight,
algorithm suggests the number of product units to be fragility, volume and other real-world constraints.
ordered and a worker in charge of stock planning is
analysing the suggestion giving us a useful feedback. REFERENCES
Based on the current results, it is estimated that stock [1] A. M. Atieh et al., “Performance Improvement of Inventory
supplies can be reduced by 5% immediately up to 10% in Management System Processes by an Automated Warehouse
Management System,” in Procedia CIRP, 2016.
the next 6 months. [2] R. Pulungan, S. P. Nugroho, N. El Maidah, T. B. Atmojo, P. D.
The algorithm for vehicle routing is using well known Hardo, and P. Pawenang, “Design of an Intelligent Warehouse
heuristic methods in each phase. On instances where the Management System,” in Information Systems International
algorithm is tested, the route proposals were Conference (ISICO), 2013.
[3] B. Yan, C. Yiyun, and M. Xiaosheng, “RFID technology applied
approximately 2% different from the optimal routes on in warehouse management system,” in Proceedings - ISECS
average. This percentage depends mostly on the number International Colloquium on Computing, Communication, Control,
of customers to be served, topographic layout of the and Management, CCCM 2008, 2008.
cities, established customer and vehicle constraints and [4] C. K. M. Lee, Y. Lv, K. K. H. Ng, W. Ho, and K. L. Choy,
“Design and application of Internet of things-based warehouse
fleet of vehicles available at a given time. Routes were management system for smart logistics,” International Journal of
created for on average over 120 customers in several Production Research. 2017.
different cities. Most customers had defined time [5] A. Giuseppe, E. Mario, and M. Cinzia, “Analysis of a warehouse
windows and the fleet consisted of less than 10 available management system by means of simulation experiments,” in 10th
International Conference on Modeling and Applied Simulation,
heterogeneous vehicles on average. Result quality was MAS 2011, Held at the International Mediterranean and Latin
influenced by the fact that with the extremely complex set American Modeling Multiconference, I3M 2011, 2011.
of input data the primary purpose of the algorithm was to [6] F. Petropoulos, S. Makridakis, V. Assimakopoulos, and K.
find any solution that would meet the set limits, all in a Nikolopoulos, “‘Horses for Courses’ in demand forecasting,” Eur.
J. Oper. Res., 2014.
reasonable time interval. [7] E. Zunic, H. Hasic, K. Hodzic, S. Delalic, and A. Besirevic,
The algorithm routes were compared to routes created “Predictive Analysis based Approach for Optimal Warehouse
manually by the managers with many years of experience. Product Positioning,” 41st International Convention on
It is noticed that in cases where the number of customers Information and Communication Technology, Electronics and
Microelectronics (MIPRO), 2018.
is about 100, the algorithm uses at least one vehicle less [8] E. Zunic, A. Besirevic, R. Skrobo, H. Hasic, K. Hodzic, and A.
than the manager. The implemented algorithm reduces Djedovic, “Design of optimization system for warehouse order
costs by approximately 30% for the mentioned number of picking in real environment,” in ICAT 2017 - 26th International
customers. The main reasons for the cost savings are the Conference on Information, Communication and Automation
Technologies, Proceedings, 2017.
better usage of smaller vehicles, a larger number of [9] E. Zunic, K. Hodzic, H. Hasic, R. Skrobo, A. Besirevic, and D.
articles in used vehicles, as well as an accumulatively Donko, “Application of advanced analysis and predictive
reduced travelled distance. In cases where the number of algorithm for warehouse picking zone capacity and content
customers is larger than 100, the savings are getting closer prediction,” in ICAT 2017 - 26th International Conference on
Information, Communication and Automation Technologies,
to 40%. Proceedings, 2017.
[10] E. Zunic, A. Besirevic, S. Delalic, K. Hodzic, and H. Hasic, “A
V. CONCLUSION Generic Approach for Order Picking Optimization Process in
Different Warehouse Layouts,” 41st International Convention on
The warehouse management system is one of the most
Information and Communication Technology, Electronics and
important parts of the working process in distribution Microelectronics (MIPRO), 2018.
companies. The optimized processes can make large time [11] O. Vallis, J. Hochenbaum, and A. Kejariwal, “A Novel Technique
and cost savings and can make the more efficient for Long-term Anomaly Detection in the Cloud,” Proccedings 6th
USENIX Work. Hot Top. Cloud Comput. (USENIX ’14), 2014.
environment. The concept described in this paper is using
the artificial intelligence algorithms to improve standard

You might also like