A simulation-based experimental design for SBS/RS warehouse design by considering energy related performance metrics

doi:10.1016/j.simpat.2019.101991

Simulation Modelling Practice and Theory

Volume 98, January 2020, 101991

https://doi.org/10.1016/j.simpat.2019.101991 Get rights and content

Highlights

•
A DOE is implemented for SBS/RS warehouse design.
•
Significant factors affecting SBS/RS warehouse performance are identified.
•
The best levels of those significant factors are determined.
•
The SBS/RS warehouse performance is also evaulated by energy related performance measures.

Abstract

This paper deals with an experimental study for shuttle-based storage and retrieval system (SBS/RS) design to identify significant factors affecting the performance metrics in the system. SBS/RS is an automated warehouse technology developed to overcome recent order profile resulting from increased e-commerce. Recent order profile has been shaped towards more product variety with less volume and short response time. In SBS/RS technology, each tier of an aisle typically has a dedicated shuttle processing orders hence, it has a very high transaction rate compared to its opponents. Neverthless, there can be several design options that can improve the performance of these systems. In this study first, the significant design factors affecting the pre-defined SBS/RS performance measures (e.g., average cycle time and energy consumption for a transaction; amount of energy regeneration for a transaction, etc.) are identified. Secondly, the best levels of these critical design factors improving the performance measures are identified. For this purpose, a statistical test, namely Tukey's test, is utilized. For the modelling purpose, simulation modelling using ARENA 14.0, is utilized. The statistical analyses from the simulation results are performed using Minitab 17 statistical software.

Introduction

Recently, utilization of robotic technologies in industries has gained popularity due to recent technological and Industry 4.0 developments. Distribution centers are also eager to automate their facilities to achieve high performance in their supply chains. For instance, Amazon, DHL, Walmart, and some other companies are automating their warehouses at an increased rate. Especially, due to the recent increase in e-commerce, warehouses prefer utilizing those robotic technologies. One of those new automated warehousing technologies is shuttle-based storage and retrieval system (SBS/RS) that is developed as an alternative to mini-load crane-based Automated Storage and Retrieval System (AS/RS) to alter low order size with short response time [2], [8], [12], [13], [14]. In 2016, the European Materials Handling Federation (FEM) reported an order statistics on the product groups of intra-logistics systems that is shown in Fig. 1. It can be observed that the order amount for SBS/RS is doubled in 2016, compared to the previous years. This increase may continue constantly due to incerase in e-commerce activities.

Typically an SBS/RS design comprises multiple tiers of storage with dedicated shuttles for each level. Shuttles provide horizontal movement for loads within a tier. These are robotic order pickers that also complete storage of mini loads (i.e., totes) in racks. This automated storage technology has the ability to process high transaction rate in which a more “traditional” mini-load AS/RS crane may be inadequate for that transaction rate (see for example, Figs. 2a, b and 3, http://www.dematic.com/multishuttle, accessed on 13/09/2019).

Lifting devices (i.e., elevators) installed at the end of each aisle enables the transfer of mini-loads between tiers. As there is a dedicated shuttle in each tier and a single lifting mechanism in each aisle, lifts may become a cause of delay in the system. Hence, a lifting mechanism with dual table design enabling process of two totes independently is developed to decrease this delay. In such a system with dual lifting mechanism, there are two buffer locations in each tier where totes are dropped-off and picked up by the shuttles and lifts. Fig. 2 shows such a design. Fig. 3 shows top view of an SBS/RS in which the shuttle drops-off/picks up a load.

İt is important to invest on the right design of the SBS/RSs at first because it affects the efficiency of the system. For instance, how many aisles, tiers and bays as well as what velocity profiles in terms of maximum speed that lifts and shuttles can reach and acceleration/deceleration values of them would provide the desired performance levels, etc. should be clear, before a company invests on these systems. These questions are also critical for the cost information of these systems. To summarize, this paper is the first to investigate the significant design factors affecting the pre-defined performance measures from an SBS/RS including energy consumption and regeneration metrics. To achieve this, a simulation-based Design of Experiment (DOE) solution approach is implemented. Then, to find out the best levels of those identified significant design factors, a statistical test, Tukey's test, is applied.

Section snippets

Literature review

The studies on SBS/RS are described in this section. Marchet et al. [14] presented an analytical model to estimate SBS/RS performance measures (the transaction cycle time and waiting times). The model is based on an open queuing network approach. The model effectiveness in performance estimation is validated through simulation. Later, Marchet et al. [15] use simulation to highlight main design trade-offs for SBS/RS for several warehouse design scenarios involving tier-captive shuttle carriers.

Simulation modelling of the sbs/rs

In the simulation model, the SBS/RS warehouse is assummed that it consists of racks, lifts, and shuttles. Two types of transactions, storage and retrieval, are processed whose process details are presented by Fig. 4. Shuttles are tier-captive and their main purpose is to store and retrieve totes from their storage locations. Racks (on either side of an aisle) consist of bays, each of which can hold one tote at a time. Because each aisle of the studied SBS/RS is identical in terms of the number

Velocity versus time graphs for travel time calculations

In the simulation model, since the average cycle time of a transaction is one of the performance measures considered in the paper, in this section, computations for travel time of a transaction is presented [4]. In travel time calculation, it is important to know whether the lift/shuttle is accelerating, decelerating or travelling at a constant velocity (maximum speed). Fig. 5a, b represent the time versus velocity graphs. Fig. 5a is the case that the travel distance is short so that

Calculations for energy consumption of a transaction

In this section, average energy consumption calculations for a transaction, W, in the studied SBS/RS is shown. The calculations are summarized from Ekren et al. [4] for lifts and shuttles separately.

Amount of energy regeneration calculations

The energy regeneration is a regenerative break which is a kinetic energy recovery mechanism for recovering a braking shuttle's/lift's kinetic energy by converting it to another form i.e., electricity. The recovered energy is usually stored in high voltage batteries for later usage. The related calculations are also summarized from Ekren et al. (2018) below.

Experimental design

In this section, the experimental design study is presented. The aim of using this technique is to identify significant factors affecting the performance of SBS/RS as well as to find out the best levels of them improving the pre-defined performance measures. Experimental design or DOE is a design tool analyzing the relation between independent (i.e., design factors) and dependent (i.e., performance measures) variables in order to identify the significant factors affecting the dependent

Conclusion

In this paper, a simulation-based Design of Experiment (DOE) study is completed for design of an SBS/RS warehouse. Firstly, performance measures (e.g., average cycle time and energy consumption for a transaction; amount of energy regeneration for a transaction) and design factors (e.g., velocity profiles of lifts and shuttles and rack design) that could affect those performance measures are defined for the studied SBS/RS. Secondly, DOE table is developed by assigning different levels for these

Acknowledgment

This work was supported by The Scientific and Technological Research Council of Turkey and Slovenian Research Agency: ARRS [grant number: 214M613].

References (21)

H.J. Carlo et al.
Sequencing dynamic storage systems with multiple lifts and shuttles
Int. J. Product. Econ.
(2012)
B.Y. Ekren et al.
A tool for time, variance and energy related performance estimations in a shuttle-based storage and retrieval system
Appl. Math. Model.
(2018)
B.Y. Ekren et al.
Simulation-based performance analysis of autonomous vehicle storage and retrieval system
Simul. Model. Pract. Theory
(2011)
Y. Ha et al.
Free balancing for a shuttle-based storage and retrieval system
Simul. Model. Pract. Theory
(2018)
Z. Ning et al.
An efficient simulation model for rack design in multi-elevator shuttle-based storage and retrieval system
Simul. Model. Pract. Theory
(2016)
D. Roy et al.
A multi-tier linking approach to analyze performance of autonomous vehicle-based storage and retrieval systems
Comput. Oper. Res.
(2017)
B. Zou et al.
Modeling parallel movement of lifts and vehicles in tier-captive vehicle-based warehousing systems
Eur. J. Oper. Res.
(2016)
M. Borovinšek et al.
Multi-Objective optimization model of shuttle-based storage and retrieval system
Transport
(2017)
B.Y. Ekren
Graph-based solution for performance evaluation of shuttle-based storage and retrieval system
Int. J. Product. Res.
(2017)
B.Y. Ekren et al.
A performance calculator for shuttle-based storage and retrieval system design

There are more references available in the full text version of this article.

Cited by (44)

Joint green dynamic order batching and picker routing problem using PSO with global worst experience
2024, Applied Soft Computing
With the increasing importance of various countries’ initiatives on reducing carbon emissions (e.g., carbon border adjustment mechanism, CBAM), intelligent order picking systems have assisted electronic retailers in realizing a green supply chain. Previous studies on picker routing operations focused on adopting offline static customer order information to carry out operational decisions. However, in practice, the customer order information is updated dynamically; and high-efficiency warehouse layouts have been increasingly concerned. Therefore, this study creates a mixed-integer programming model for minimizing both the total carbon footprint and the total penalty cost for delayed orders in the joint order batching and picker routing problem with dynamic arriving orders and shipping time constraints in a high-efficiency fishbone warehouse layout. To solve this complex problem, a particle swarm optimization algorithm that integrates the swarm’s previous global and local worst experiences and a migration mechanism is further proposed to increase solution quality and computing efficiency. In addition, a three-dimensional space for the warehouse floorplan and shipping time constraints is designed to divide orders into batches to address the constraints. Theoretical analysis of this algorithm is conducted, and experimental analysis shows that this algorithm finds superior solutions than current practical strategies, to offer a valuable reference for carbon emission reduction in greening warehousing.
Review of Relevant Literature on Modelling and Simulation Approaches for AS/RSs
2023, IFAC-PapersOnLine
Automated Storage and Retrieval Systems support warehouse processes to improve production and logistics efficiency, reducing operating times and costs. In recent years, much attention has been paid on evaluating the performance of these systems using different simulation and modelling approaches. However, the current body of literature is quite fragmented and there is the need for a holistic overview of published scientific literature. To this end, this work presents a literature review in order to adequately identify relevant scientific trends and, consequently, the main research gaps in dealing with the systems under study. From the analysis of the obtained corpus of papers it can be observed that Shuttle Based Storage and Retrieval Systems are the most studied solutions, the most adopted approach is Discrete Event Simulation and operational impacts are the most frequently considered. By focusing on the research gaps, it can be highlighted that the validation process of the chosen methodology is often overlooked and the economic and ergonomic impacts are barely addressed.
Retrieval request scheduling in a shuttle-based storage and retrieval system with two lifts
2023, Transportation Research Part E: Logistics and Transportation Review
We study a shuttle-based storage and retrieval system with two independent lifts shared in a single mast. In such a system, two lifts move simultaneously but cannot pass each other, and each retrieval request is finished with the cooperation of a lift and a shuttle. Therefore, given a set of retrieval requests, the retrieval request scheduling problem consists of determining the sequence in which the requests are retrieved and assigning each request to a lift with the objective of minimizing the makespan, considering both the lift-lift interaction and lift-shuttle interaction. The problem is formulated as a mixed-integer programming model and proved to be NP-hard. We propose a decomposition-based adaptive large neighborhood search heuristic to quickly compute near-optimal solutions, using the property that the assignment of requests to lifts for a given retrieval sequence can be exactly solved in polynomial time by a dynamic programming approach. Numerical results indicate that our algorithm achieves a lower makespan than the methods proposed in the literature and used in practice. We also apply the proposed algorithm using real data to a realistic setting that considers multiple planning horizons, showing that it significantly outperforms the policy the company currently uses.
Travel time models for tier-to-tier SBS/RS with different storage assignment policies and shuttle dispatching rules
2021, Transportation Research Part E: Logistics and Transportation Review
Citation Excerpt :
Instead of cranes, a shuttle-based storage and retrieval system (SBS/RS) uses lifts and shuttles as S/R machines to improve flexibility and productivity. In an SBS/RS, illustrated by Fig. 1, each aisle is equipped with a lift for vertical movements (Ekren, 2017, 2020; Küçükyaşar et al., 2021). The rack system is divided into tiers; and aisle-captive shuttles (Eder, 2020a; Lerher et al., 2021) are used to move SKUs on each tier.
This paper explores both random and class-based storage policies and three shuttle dispatching rules, random, distance-based, and demand-rate based, for a tier-to-tier shuttle-based storage and retrieval system (SBS/RS). Modeling the system as a discrete-time Markov Chain, this study derived the shuttle distribution under each policy combination, and further developed the expected travel time models for both single-command (SC) and dual-command (DC) operations. The developed models can help a warehouse designer or manager to comprehensively make four decisions for SBS/RS, whether SC or DC operations should be adopted, the most appropriate storage policy, the best shuttle dispatching rule, and the number of shuttles. The travel time models can evaluate each possible policy combination quickly through multiple criteria, such as the capacity, customer responsiveness, and costs of shuttles. Numerical experiments showed significant impacts of the policy combination on the expected travel time. Class-based storage, if implementable, is always better than random storage under each combination of shuttle dispatching rule and operational mode (i.e., SC or DC). Under random storage and SC operations, the distance-based shuttle dispatching rule is always better than the random rule, and their difference is concave in the number of shuttles. When classed-based storage is adopted and the number of shuttles is small, the demand-rate-based rule is the best under SC operations. Its benefit becomes more obvious when demand rates become more heterogeneous until some point but decreases as the system has more shuttles. Under DC Operations, depending on equipment speeds, the distance-based shuttle dispatching rule can be worse than the random rule under random storage. Under class-based storage and DC operations, the demand-rate-based rule is still the best when demand rates are heterogeneous. The differences between SC and DC operations are more significant when demand rates are more homogenous. When a system wants to be more responsive to customers, it should be run under SC operations. If the overall capacity is of interest, DC operations are much better.
A method for developing and validating simulation models for automated storage and retrieval system digital twins
2024, International Journal of Advanced Manufacturing Technology
Scheduling Optimization of Compound Operations in Autonomous Vehicle Storage and Retrieval System
2024, Symmetry

View all citing articles on Scopus

View full text

A simulation-based experimental design for SBS/RS warehouse design by considering energy related performance metrics

Highlights

Abstract

Introduction

Section snippets

Literature review

Simulation modelling of the sbs/rs

Velocity versus time graphs for travel time calculations

Calculations for energy consumption of a transaction

Amount of energy regeneration calculations

Experimental design

Conclusion

Acknowledgment

Int. J. Product. Econ.

Appl. Math. Model.

Simul. Model. Pract. Theory

Simul. Model. Pract. Theory

Simul. Model. Pract. Theory

Comput. Oper. Res.

Eur. J. Oper. Res.

Multi-Objective optimization model of shuttle-based storage and retrieval system

Transport

Graph-based solution for performance evaluation of shuttle-based storage and retrieval system

Int. J. Product. Res.

A performance calculator for shuttle-based storage and retrieval system design