Stowage Planning Research Articles

Lashing Force Prediction Model with Multimodal Deep Learning and AutoML for Stowage Planning Automation in Containerships

The calculation of lashing forces on containerships is one of the most important aspects in terms of cargo safety, as well as slot utilization, especially for large containerships such as more than 10,000 TEU (Twenty-foot Equivalent Unit). It is a challenge for stowage planners when large containerships are in the last port of region because mostly the ship is full and the stacks on deck are very high. However, the lashing force calculation is highly dependent on the Classification society (Class) where the ship is certified; its formula is not published and it is different per each Class (e.g., Lloyd, DNVGL, ABS, BV, and so on). Therefore, the lashing result calculation can only be verified by the Class certified by the Onboard Stability Program (OSP). To ensure that the lashing result is compiled in the stowage plan submitted, stowage planners in office must rely on the same copy of OSP. This study introduces the model to extract the features and to predict the lashing forces with machine learning without explicit calculation of lashing force. The multimodal deep learning with the ANN, CNN and RNN, and AutoML approach is proposed for the machine learning model. The trained model is able to predict the lashing force result and its result is close to the result from its Class.

Optimizing storage location assignment in an automotive Ro-Ro terminal

Automotive roll-on/roll-off (Ro-Ro) transportation is an efficient and competitive method for the large-scale transshipment of commercial cars. However, the low-efficiency operations and insufficient storage resources of automotive Ro-Ro terminals have constrained the development of Ro-Ro transportation. This paper investigates the storage location assignment problem (SLAP) for the arrival of cars at the yard, and it aims to improve the ship-loading efficiency and contribute to efficient storage at Ro-Ro terminals. Two deadlock situations resulting from blocked routes are analyzed in detail. Based on the Ro-Ro ship stowage plan, a car group concept is proposed to reflect the loading sequence of cars into a Ro-Ro ship. The dispersion degree is defined to describe the centralized layout of every car group in the yard. A linear 0-1 integer programming model is formulated to minimize the total dispersion degrees of all car groups. Furthermore, an indicator called the attraction degree is presented to quantify the preferred degree of each location for storing different groups of cars. A hierarchical two-stage exchange strategy (HTSES) is designed to obtain the car layout with the minimum total dispersion degree. To reduce the scale of the solved problem, a rolling-horizon heuristic approach based on closed-loop (positive and negative) feedback is proposed. Positive feedback based on the guidance mechanism describes the guidance provided by the existing car layout to arriving cars, while negative feedback based on the reformulation mechanism represents the influence of arriving cars on the car layout. A series of numerical experiments show that the proposed method can effectively produce a satisfactory car assignment plan for the management of automotive Ro-Ro terminals.

New uncertainty modelling for cargo stowage plans of general cargo ships

The current approach to the cargo stowage plans (CSP) of general cargo ships (GCS) is safety-driven, which means that any CSP satisfying minimum safety requirements can be used in practice. Such an approach taking into account no economic and environmental concerns cannot help sustain GCS growth in today’s competitive freight transportation market. This paper introduces a revised evidential reasoning (ER) approach to cope with the complex decision-making problem associated with the CSP of GCS. The complexity mainly results from the dynamic interdependency between the decision criteria and alternatives. The revised ER can determine the functions of the safety-related criteria in the decision making process by considering the extent to which each decision alternative meets the minimum safety requirements. The model is tested in multiple forms by an empirical study using a national GSC loading laboratory, and a real-life application by a shipping company in practice. The results reveals that the new model can aid general ship owners to make sustainable CSPs from a multiple-dimensional perspective and select an optimal CSP based on specific voyage scenarios.

A heuristic and a benchmark for the stowage planning problem

The stowage planning problem has recently gained the attention of a number of academic researchers. Unfortunately, many of the published works are either based on oversimplified assumptions or on confidential data. This practice hinders the research field from growing. In this paper, we present a novel set of realistic vessel data along with a set of benchmark instances. Moreover, a formal definition of a single-port stowage planning problem, based on the current state of the art, is presented. The proposed optimisation problem is solved using a variant of the adaptive large neighbourhood search framework, where novel repair and destroy methods are presented. Computational results show that the solution approach is able to find high-quality seaworthy stowage plans within 60 s.

Automatic Stowage of Bulk Carrier Based on Ship Floating-State Control

The quality of stowage plan affects several aspects of ship security, including intact stability and structural strength. This study focuses on optimization of the stowage plan under a given floating state (i.e., the final heel and trim angle). The mathematical model for automatic stowage plan is built by using the following objectives: minimizing the total ballast water, maximum shear force (SF), and bending moment (BM), and maximizing the total cargo. The related safety assessment criteria, the final heel and trim angle, maximum tank capacity, and amount of cargo are considered as constraints. A two-layer optimization method based on differential evolution algorithm is proposed to solve the constrained optimization problem. According to the characteristics of bulk carriers, a simplified matrix method for engineering applications is discussed based on Newton’s iteration method to calculate ship floatation. To improve the calculation accuracy of SF and BM, a method based on the stereolithography model is proposed. Finally, an actual case of a 38,500-dead weight tonnage (DWT) bulk carrier is used to illustrate the effectiveness of the proposed approach.

Efficient stowage plan with loading and unloading operations for shipping liners using foldable containers and shift cost-sharing

ABSTRACT In this paper, we investigate the foldable container slot planning problem with loading and unloading operations that include shifting containers in a shipping line. We use the global optimal perspective in which a terminal operator generates an optimal stowage plan created on the basis of demand at subsequent ports. State-of-the-art foldable containers have been recently used in commercial maritime transport systems because they confer space-saving advantages when folded. We investigate container use through mixed-integer programming and shift cost-sharing methods as means to prevent conflicts between ports over inessential shifts and to provide guidelines for distributing shift costs among all ports in a logical and fair way. Through the proposed model, we found that most inessential shifts, considered inevitable from the local optimal perspective, can be eliminated, and the inevitable shift costs can be fairly distributed. This article is a revised and expanded version of a paper entitled ‘Shift Minimization with Loading and Unloading Operations using Foldable Containers’ presented at the first Conference of the Yangtze-River Research and Innovation Belt(Y-RIB), Zhoushan, China; 2–5 December 2018.

An adaptive large neighborhood search heuristic for the planar storage location assignment problem: application to stowage planning for Roll-on Roll-off ships

This paper considers a generalized version of the planar storage location problem arising in the stowage planning for Roll-on/Roll-off ships. A ship is set to sail along a predefined voyage where given cargoes are to be transported between different port pairs along the voyage. We aim at determining the optimal stowage plan for the vehicles stored on a deck of the ship so that the time spent moving vehicles to enable loading or unloading of other vehicles (shifting), is minimized. We propose a novel mixed integer programming model for the problem, considering both the stowage and shifting aspect of the problem. An adaptive large neighborhood search (ALNS) heuristic with several new destroy and repair operators is developed. We further show how the shifting cost can be effectively evaluated using Dijkstra’s algorithm by transforming the stowage plan into a network graph. The computational results show that the ALNS heuristic provides high quality solutions to realistic test instances.

Loading Sequencing Problem in Container Terminal with Deep Q-Learning

Xia, M.; Li, Y.; Shen, Y., and Zhao, N., 2020. Loading sequencing problem in container terminal with Deep Q-Learning. In: Yang, Y.; Mi, C.; Zhao, L., and Lam, S. (eds.), Global Topics and New Trends in Coastal Research: Port, Coastal and Ocean Engineering. Journal of Coastal Research, Special Issue No. 103, pp. 817–821. Coconut Creek (Florida), ISSN 0749-0208.Loading sequencing is a scheduling problem in container terminal to reduce handling cost and improve handling efficiency of loading process by rearranging the loading sequencing preplanned by stowage planning. Loading sequencing orchestrates terminal equipment to handle export container loading efficiently. To obtain good solving efficiency in large scale scheduling cases, a double Deep Q-Network with prioritized experience replay is proposed to solve loading sequencing problem. In this approach, a simulation with different complexity scale is introduced to emulate the loading process. With real world data training, the proposed PERDDQN can solve loading sequencing episode in seconds. Result shows good efficiency and generalization of proposed algorithm. Simple simulation trained network evaluated by complex simulation shows that the proposed approach could be applicable to real word scheduling.

Design and application of reuse stage algorithm of case-based reasoning method on container stowage planning

Stowage planning is a container arrangement planning activity on container ships. The Case-Based Reasoning method can solve the newest stowage plan cases by taking a stowage plan case that has the highest similarity and has been resolved in the past as a reference for arranging. The Case-Based Reasoning method has four-stage, there are Retrieve, Reuse, Revise, Retain. The Reuse stage is to adapt the Container Loading List so that it is arranged to resemble the case that was taken at the Retrieve stage. Algorithms for the Reuse stage of the Case-Based Reasoning Method has been designed and obtained the Container Loading List Adaptation, Fill-Blank, Adding the Remaining Container, Trim Optimization, and GM Optimization Functions. The functions can be implemented in software design and automatically or manually arranged the containers based on Casebase. Hence, users make the new stowage plans easier and faster. Based on the results of software testing, the difference between LCB and LCG is less than 1% LBP so that the Trim was fulfilled, and GM obtained a positive value so that the ship was in a stable condition.

Integrated optimisation of consolidation and stowage planning of steel coil ships using differential evolution

Consolidation plan (CP) and stowage plan (SP) of coil ships are to assign steel coils onto ships and decide specific loading location for each coil, respectively, which are two correlative and adjacent plans in the finished product logistics of iron and steel industry. In this paper, we focus on the problem of making an integrated plan of CP and SP (C&SPP), which is important and rarely studied, to try to achieve systematic optimisation from the perspectives of the product terminal, shipping companies, and customers, simultaneously. First, according to the practical situation, an integer programming (IP) model is established to optimise ship loading, transportation timeliness, and operation efficiency. Then, to improve calculating efficiency, the IP model is simplified (SIP) by reducing the dimensions of variables. After that, based on the problem features and algorithm behaviour, an improved differential evolution algorithm is proposed in which an element-dependent strategy is designed to set control parameter, a guided operator is constructed to generate mutant individual and an iteration-based mechanism is introduced to dynamically adjust the parameter and mutation strategy, respectively. Finally, extensive experiments are carried out to evaluate the proposed method, and the numerical results demonstrate its effectiveness in solving practical scale C&SPP.

Optimizing the Stowage Planning and Block Relocation Problem in Inland Container Shipping

The stowage planning and block relocation problem (SPBRP) consider the relations between stowage planning and block relocation comprehensively, in which the retrieval sequence of containers in yard and the stowage plan of a ship are concerned. In this paper, an integer programming formulation called SPBRP-I is presented for the proposed problem in inland container liner shipping to minimize the total number of relocations in yard and re-handles in ship. To deal with the utilization preference in inland shipping, the minimization of ship stack occupancy number is introduced into SPBRP-I to develop a new model called SPBRP-II. In view of the complexity of problem-solving, a heuristic algorithm (HA) based on priority sorting, retrieval and local search strategies is proposed to achieve the efficient optimization. The algorithm consists of three parts including the yard container retrieving, yard container relocation and ship stowage planning. Finally, the container liner shipping on the Yangtze River is chosen with some typical ships for the case study. Numerical results show that SPBRP-II performs better than the original SPBRP-I, and it's more suitable for solving the small-sized cases. Compared with the models, the algorithm with retrieval strategy RS1 (HA_RS1) can achieve the efficient problem solving with high-quantity solutions and very short computational time.

Solving inland container ship stowage planning problem on full route through a two-phase approach

Inland container ship emphasises capacity utilisation due to its limited capacity. Its stability is very sensitive to the stowage plan. The shipping line planners are under pressure to make efficient stowage plan on full route. This paper adopts a two-phase approach to separate the problem into two planning levels: multi-port master bay plan problem (MP-MBPP) on full route and slot plan problem (SPP) for each bay at each port. The mathematical models are proposed respectively for both sub-problems based on mixed integer programming. For each sub-problem, the greedy randomised adaptive search procedure (GRASP) and the heuristic evolutionary strategy (HES) algorithm are proposed, respectively. The grey entropy parallel analysis (GEPA) method is presented to guide HES in multi-objective optimisation. Experimental results based on real-world scenarios are presented to show the efficacy of proposed algorithms and method.

Solving inland container ship stowage planning problem on full route through a two-phase approach

Inland container ship emphasises capacity utilisation due to its limited capacity. Its stability is very sensitive to the stowage plan. The shipping line planners are under pressure to make efficient stowage plan on full route. This paper adopts a two-phase approach to separate the problem into two planning levels: multi-port master bay plan problem (MP-MBPP) on full route and slot plan problem (SPP) for each bay at each port. The mathematical models are proposed respectively for both sub-problems based on mixed integer programming. For each sub-problem, the greedy randomised adaptive search procedure (GRASP) and the heuristic evolutionary strategy (HES) algorithm are proposed, respectively. The grey entropy parallel analysis (GEPA) method is presented to guide HES in multi-objective optimisation. Experimental results based on real-world scenarios are presented to show the efficacy of proposed algorithms and method.

A Mathematical Model and Two-Stage Heuristic for the Container Stowage Planning Problem With Stability Parameters

Owing to the significant increase in the volume of world trade, mega-container vessels are being used to meet transportation demands. As the size of the vessels increases, the loading sequence of containers onto the vessels presents an important challenge for planners. In this study, we consider the container stowage planning problem with stability constraints (e.g. shear force, bending moment, trim) and develop a mixed integer linear programming (MILP) formulation which generates load plans by minimizing total cost associated with the over-stows and trimming moments. Our study adopts a holistic perspective which encompasses several real-world features such as different container specifications, a round-robin tour of multiple ports, technical limitations related to stack weight, stress, and ballast tanks. We also propose a two-stage heuristic solution methodology that employs an integer programming (IP) formulation and then a swapping heuristic (SH) algorithm. This approach first acquires a lower bound on the total over-stow cost with the IP model, thereby creating an initial bay plan. Then, it applies the SH algorithm to this initial bay plan to minimize cost resulting from trimming moments. The efficiency of the MILP formulation and heuristic algorithm is investigated through numerical examples. The results have shown that the heuristic has greatly improved the solution times as well as the size of the solvable problems compared to the MILP formulation. In particular, the two-stage heuristic can solve all size problem instances within an average optimality gap of 0-25% in less than 8 minutes, whereas the MILP can only achieve an approximate optimality gap of 55-80% in 2 hours.

Loading tow trains ergonomically for just-in-time part supply

Faced with an aging workforce, many manufacturing companies consider alleviating the ergonomic strain of material handling on their workers increasingly important. This is one of the reasons why frequent small-lot deliveries of parts to the assembly stations on the shop floor via small electric delivery vehicles – so-called tow trains – have become widespread in many industries. Deploying tow trains, however, does not automatically ease the ergonomic burden on logistics workers, but requires careful stowage planning in addition. In this paper, we consider the following problem. Given a set of bins of differing weight to be carried by tow train to a given set of stations on the shop floor, where should each bin be stowed on the tow train such that it can be unloaded efficiently from an economic perspective while also minimizing the ergonomic strain during loading and unloading? We investigate the physiological stress of handling bins on different levels of a tow train wagon by applying an established ergonomic evaluation method from the human factors engineering literature. We model the ensuing optimization problem as a special type of assignment problem and propose suitable exact and heuristic solution methods. In a computational study, our approaches are shown to perform well, delivering optimal solutions for instances of realistic size within fractions of a second in many cases. We show that optimal stowage plans can significantly ease the physiological burden on the workforce without compromising economic efficiency. We also derive some insights into the ideal layout of the tow train from an ergonomics perspective.

Matheuristics for slot planning of container vessel bays

Stowage planning is an NP-hard combinatorial problem concerned with loading a container vessel in a given port, such that a number of constraints regarding the physical layout of the vessel and its seaworthiness are satisfied, and a number of objectives with regard to the quality of the placement are optimized. State-of-the-art methods decompose the problem into phases, the latter of which, known as slot planning, involves loading the containers into slots of a bay. This article presents an efficient matheuristic for the slot planning problem. Matheuristics are algorithms using mathematical programming techniques within a heuristic framework. The method finds solutions for 96% of 236 instances based on real stowage plans, 90% of them optimally, with an average optimality gap of 4.34% given a limit of one second per instance. This is an improvement over the results provided by previous works.

A decision-support framework for the stowage of maritime containers in inland shipping

The typical voyage of a barge transporting maritime containers inland, consists of visiting a set of terminals where loading/unloading operations occur. The required stowage plans must meet stability requirements during all transport phases and avoid costly re-handling operations. In this paper, we formalize this problem and consider a typical dry-port transport system. We propose a comprehensive mathematical model that seeks to maximize stowed containers. To solve it, we develop a hybrid metaheuristic approach, based on local search and an industrial solver. Numerical experiments, based on real-world data, provide insights into the performances of the proposed framework and current stowage practices.

An exact algorithm for the block relocation problem with a stowage plan

In this paper we address an exact algorithm for the block relocation problem with a stowage plan. This problem is an extension of the container (block) relocation problem that aims at minimizing the total number of relocations required for retrieving containers piled up in a container yard. The difference is that a stowage plan in a vessel is given in advance and it is taken into consideration when retrieving containers. More specifically, the retrieval order of containers should be feasible with respect to a given stowage plan. We construct a branch-and-bound algorithm with iterative deepening for unrestricted and restricted variants of this problem. To this end, we propose three lower bounds on the total number of relocations. We also extend dominance rules for the container relocation problem, in order to restrict the search space and improve the search efficiency. The effectiveness of the proposed algorithm is examined by numerical experiments for benchmark instances from the literature.

Container Ship Stowage Based on Monte Carlo Tree Search

ABSTRACT Zhao, N.; Guo, Y.; Xiang, T.; Xia, M.; Shen, Y., and Mi, C., 2018. Container ship stowage based on monte carlo tree search. In: Liu, Z.L. and Mi, C. (eds.), Advances in Sustainable Port and Ocean Engineering. Journal of Coastal Research, Special Issue No. 83, pp. 540–547. Coconut Creek (Florida), ISSN 0749-0208. With the development of larger ships, modernized wharves, standardized production, meticulous management, and flexible service in container transportation, port production management is becoming increasingly intelligent. How to improve the efficiency when making stowage plans, the core issue in container terminals, has become a popular topic in research. Starting with the stowage issue, this paper provides rational solutions to stowage location and loading sequence of unloaded containers on the yard, giving overall consideration on ship owners' requirement and dock authority's practical operation. Considering turnover on yard, RTG cross-container operation and RTG shifting, based on loadi...

The outbound container slot allocation based on the stowage plan in rail–water intermodal container terminals

In the past, most researchers focused on the storage space allocation problem or container block allocation problem in maritime container terminals, while few studied the container slot allocation problem in rail–water intermodal container terminals. Container slot allocation problem is proposed to reduce relocation operations of containers in railway container yards and improve the efficiency of rail–water intermodal container terminals. In this paper, a novel outbound container slot allocation model is introduced to reduce the rehandling operations, considering stowage plan, containers left from earlier planning periods and container departure time. A novel heuristic algorithm based on the rolling planning horizon approach is developed to solve the proposed problem effectively. Computational experiments are carried out to validate that the proposed model and algorithm are feasible and effective to enhance the storage effect. Meanwhile, some other experiments are conducted to verify that our approach is better than the regular allocation approach, which is a common method in marine and railway container terminals, and container weight is the most important influence factor when storing containers.