DESC:FORM 2

THE PATENTS ACT, 1970
(39 of 1970)
&
THE PATENT RULES, 2003

COMPLETE SPECIFICATION
(See Section 10 and Rule 13)

Title of invention:

METHOD AND SYSTEM FOR CONTAINER LOADING SEQUENCE GENERATION USING REINFORCEMENT LEARNING (RL)

Applicant

Tata Consultancy Services Limited
A company Incorporated in India under the Companies Act, 1956
Having address:
Nirmal Building, 9th floor,
Nariman point, Mumbai 400021,
Maharashtra, India
The following specification particularly describes the invention and the manner in which it is to be performed.

CROSS-REFERENCE TO RELATED APPLICATIONS AND PRIORITY

[001] The present application claims priority to Indian provisional specification (Title: METHOD AND SYSTEM FOR CONTAINER LOADING SEQUENCE GENERATION USING REINFORCEMENT LEARNING (RL)) No. (201821015646), filed in India on 25th of April, 2018.

TECHNICAL FIELD
[002] The disclosure herein generally relates to container loading, and, more particularly, to generating a container loading sequence.

BACKGROUND
[003] Container loading refers to the act of loading containers to available slots in a storage deck. In almost all such scenarios, parameters such number of containers, number of slots available, shape of the storage deck (and in turn position of the slots), size and shape of containers, varies. In addition to this, certain containers may have specific requirements, not just in terms of size and shape of the slots, but in terms of facilities in the slots too. For example, a container in which items that required to be kept under sub-zero temperatures, needs to be placed in a slot where appropriate cooling facilities are available.
[004] More the number of containers to be loaded, more the required effort is. Similarly, the containers may be kept in random fashion without following any specific order. In that case, effort is more. For example, assume that a container (‘target container’) to be placed in a particular slot is kept under a few other containers. In that case to pick the target container, all the containers on top of the target container have to be moved first, which in turn adds to additional effort and time consumed. Various scheduling mechanisms exist which aid in scheduling pick and place of containers in appropriate slots. However, some of the existing scheduling mechanisms use static information, and hence are not capable of catering dynamic requirements. Certain other scheduling mechanisms generate loading sequence, however, approach used are different, and may have corresponding impact on the results. In addition to this, such mechanisms follow certain defined approaches to process dynamic data and to generate recommendations/suggestions, which means the results generated do not ‘improve’ over a period of time.

SUMMARY
[005] Embodiments of the present disclosure present technological improvements as solutions to one or more of the above-mentioned technical problems recognized by the inventors in conventional systems. For example, in one embodiment, a processor implemented method for generating recommendations for container loading is provided. In this method, a plurality of containers (Ci) corresponding to each of a plurality of slots (Si) are identified via one or more hardware processors. Further, a plurality of container loading recommendations are generated based on the plurality of containers identified for each of the plurality of slots, via the one or more hardware processors. Further, a shuffle count value is determined for each of the plurality of container loading recommendations, via the one or more hardware processors, wherein the shuffle count value of a container-slot pair in the plurality of container loading recommendations represents number of other containers to be moved so as to pick the container to load to the slot, at a time instance ti. Further, one of a positive rating or a negative rating is provided for the at least one container loading recommendation, based on corresponding shuffle count value, via the one or more hardware processors. Then each container is picked as a target container for each of the plurality of slots, via the one or more hardware processors, wherein the target container has least value of the shuffle count among all the containers having the positive rating from the plurality of containers corresponding to each slot.
[006] In another aspect, a system for generating recommendations for container loading is provided. The system includes a memory module storing a plurality of instructions, one or more communication interfaces, and one or more hardware processors. The one or more hardware processors is coupled to the memory module via the one or more communication interfaces, wherein the one or more hardware processors are caused by the plurality of instructions: to identify a plurality of containers (Ci), corresponding to each of a plurality of slots (Si); generate a plurality of container loading recommendations based on the plurality of containers identified for each of the plurality of slots; determine a shuffle count value for each of the plurality of container loading recommendations, wherein the shuffle count value of a container-slot pair in the plurality of container loading recommendations represents number of other containers to be moved so as to pick the container to load to the slot, at a time instance ti; provide one of a positive rating or a negative rating for the at least one container loading recommendation, based on corresponding shuffle count value; and pick each container, as a target container for each of the plurality of slots, wherein the target container has least value of the shuffle count among all the containers having the positive rating from the plurality of containers corresponding to each slot.
[007] In yet another aspect, a non-transitory computer readable medium for generating recommendations for container loading is provided. The non-transitory computer readable medium executes the following method for generating the recommendations. In this method, a plurality of containers (Ci) corresponding to each of a plurality of slots (Si) are identified via one or more hardware processors. Further a plurality of container loading recommendations are generated based on the plurality of containers identified for each of the plurality of slots, via the one or more hardware processors. Further, a shuffle count value is determined for each of the plurality of container loading recommendations, via the one or more hardware processors, wherein the shuffle count value of a container-slot pair in the plurality of container loading recommendations represents number of other containers to be moved so as to pick the container to load to the slot, at a time instance ti. Further, one of a positive rating or a negative rating is provided for the at least one container loading recommendation, based on corresponding shuffle count value, via the one or more hardware processors. Then each container is picked as a target container for each of the plurality of slots, via the one or more hardware processors, wherein the target container has least value of the shuffle count among all the containers having the positive rating from the plurality of containers corresponding to each slot.
[008] It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed.

BRIEF DESCRIPTION OF THE DRAWINGS
[009] The accompanying drawings, which are incorporated in and constitute a part of this disclosure, illustrate exemplary embodiments and, together with the description, serve to explain the disclosed principles:
[010] FIG. 1 illustrates an exemplary system for generating loading sequence and recommendations for container loading, according to some embodiments of the present disclosure.
[011] FIG. 2 is a flow diagram depicting steps involved in the process of generating container loading sequence and recommendations, using the system of FIG. 1, according to some embodiments of the present disclosure.
[012] FIG. 3 is a flow diagram depicting steps involved in the process of rating container loading recommendations, using the system of FIG. 1, according to some embodiments of the present disclosure.
[013] FIG. 4 is an example of state space representation of container and slot data, according to some embodiments of the present disclosure.

DETAILED DESCRIPTION OF EMBODIMENTS
[014] Exemplary embodiments are described with reference to the accompanying drawings. In the figures, the left-most digit(s) of a reference number identifies the figure in which the reference number first appears. Wherever convenient, the same reference numbers are used throughout the drawings to refer to the same or like parts. While examples and features of disclosed principles are described herein, modifications, adaptations, and other implementations are possible without departing from the spirit and scope of the disclosed embodiments. It is intended that the following detailed description be considered as exemplary only, with the true scope and spirit being indicated by the following claims.
[015] Referring now to the drawings, and more particularly to FIG. 1 through FIG. 4, where similar reference characters denote corresponding features consistently throughout the figures, there are shown preferred embodiments and these embodiments are described in the context of the following exemplary system and/or method.
[016] FIG. 1 illustrates an exemplary system for generating loading sequence and recommendations for container loading, according to some embodiments of the present disclosure. In an embodiment, the system 100 includes one or more memory module(s) 101, one or more hardware processor(s) 102, and one or more communication interface(s) 103. The one or more hardware processor(s) 102 are operatively coupled to the one or more memory module(s) 101.
[017] The one or more hardware processors 102 can be implemented as one or more microprocessors, microcomputers, microcontrollers, digital signal processors, central processing units, state machines, graphics controllers, logic circuitries, and/or any devices that manipulate signals based on operational instructions. Among other capabilities, the processor(s) are configured to fetch and execute computer-readable instructions stored in the memory. Execution of the computer readable instructions causes the one or more hardware processors 102 to perform steps depicted in FIG. 2 for the container loading sequence generation purpose. In an embodiment, the system 100 can be implemented in a variety of computing systems, such as laptop computers, notebooks, hand-held devices, workstations, mainframe computers, servers, a network cloud and the like.
[018] The communication interface(s) 103 can include a variety of software and hardware interfaces, for example, a web interface, a graphical user interface, and the like and can facilitate multiple communications within a wide variety of networks N/W and protocol types, including wired networks, for example, LAN, cable, etc., and wireless networks, such as WLAN, cellular, or satellite. In an embodiment, the I/O interface device(s) can include one or more ports for connecting a number of devices to one another or to another server.
[019] The memory module(s) 101 may include any computer-readable medium known in the art including, for example, volatile memory, such as static random access memory (SRAM) and dynamic random access memory (DRAM), and/or non-volatile memory, such as read only memory (ROM), erasable programmable ROM, flash memories, hard disks, optical disks, and magnetic tapes. In an embodiment, one or more modules (not shown) of the system 100 can be stored in the memory 102.
[020] FIG. 2 is a flow diagram depicting steps involved in the process of generating container loading sequence and recommendations, using the system of FIG. 1, according to some embodiments of the present disclosure. In an embodiment, though the steps are described in a sequential order, neither the figure nor the description intend to indicate that the steps are performed in that order. The steps may be performed in any order practical, and some of the steps may be even performed simultaneously.
[021] Each slot and each container has own characteristics. Here, for a slot, the term characteristics can refer to dimensions (length*width*height) of the slot, facilities at the slot (for example, availability heating/cooling facilities), and so on. Similarly for a container, the term ‘characteristics’ can refer to dimensions (length*width*height) of the container, data pertaining to item/product stored in the container, requirements (for example, cooling requirement) and so on. The characteristics of each slot and each container are collectively represented using a mask-id. For example, a mask-id assigned to a container is based on characteristics of that particular container. The mask-id may be in any suitable format, and may contain integers, alphabets, special characters, and a combination thereof. A database with information pertaining to characteristics and corresponding mask-ids may be maintained in the memory module 101. While considering slots and containers for loading purpose, the system 100 can refer to this database and identify characteristics of the slot and/or container being considered. By comparing mask-id of containers and slots, the system 100 identifies (202) all containers that match each of a plurality of slots. In an embodiment, each slot can have one or more matching containers, which means mask-id (and in turn characteristics) of all of the containers identified as the ‘matching containers’ match mask-ids (and in turn characteristics) of the slot. Based on the one or more containers identified as matching each of the plurality of slots, the system 100 generates (204) a plurality of container loading recommendations. Number of container loading recommendations for a slot may depend on number of containers that have been identified as matching the slot. For example, the container loading recommendation may be in (slot# - container#) format.
[022] The container loading recommendations are further processed by the system 100 using a Reinforcement Learning (RL) based mechanism in which the container loading recommendations are provided to an ‘agent’ (not shown) in the system 100. The agent determines (206) a shuffle count of each container loading recommendation. Here the term ‘shuffle count value’ for a (slot# - container#) container loading recommendation indicates/represents number of other containers to be shuffled/moved in order to pick and place a container in the corresponding slot. For example, 4 containers (A, B, C, and D) are stacked one above the other in the order A ? B ? C ? D from bottom to top. Assume that the system 100 identifies the containers A and D as matching containers for a slot X. Since A is in the bottom of the stack, in order to pick and place A in the slot X, the other containers (i.e. B, C, and D) are to be moved first, which means shuffle count value for the container A is 3. On the other hand, as D is on top of the stack, D can be directly picked without moving the other containers, which means the shuffle count value for D is 0. Here the shuffle count value of a container (and in turn of the corresponding container loading recommendation) indirectly indicates effort required for picking and placing a container in a particular slot.
[023] The system 100 then provides (208) one of a positive rating or a negative rating to each container loading recommendation. In this process, the system 100 compares the determined shuffle count value of each container loading recommendation with a threshold value of shuffle count. If the identified value of the shuffle count is lower than the threshold value of shuffle count for any container loading recommendation, then the system 100 assigns a positive rating, and for a shuffle count value exceeds the threshold value, a negative rating is provided. The ratings provided is an indicator of quality of recommendation. The agent is further configured to approve/disapprove (agree/disagree to) the container loading recommendations, at least by considering the rating provided to each of the container loading recommendations. For example, all the container loading recommendations having got the negative rating may be rejected by the system 100, and the all the container loading recommendations having got the positive rating may be approved by the agent/the system 100. Further, based on all the container loading recommendations the system 100 has approved (for all slots), one target container each for each of the plurality of slots is identified. For a slot, if only one container loading recommendation has got the positive rating, then the corresponding container is identified as the target container for that particular slot. In another embodiment, if for a particular slot more than one container loading recommendations have got the positive rating, then the container loading recommendation with the least shuffle count value may be identified by the system 100 and the corresponding container is identified as the target container. Based on the container identified for each slot, the system 100 generates a loading sequence. In an embodiment, the loading sequence is generated such that the shuffle count is minimum. In another embodiment, when the container loading is being carried out as per the loading sequence generated, the shuffle count changes dynamically. For example, in the aforementioned scenario of four containers stacked one above the other, when the container D is picked and placed in corresponding slot, the shuffle count value of the remaining containers reduces. In a possible mode of implementation the system 100 can dynamically identify the target container for each of the remaining slots, based on the shuffle count determined after the previous picking and placing of one of the containers. Also, data pertaining to the ratings, and in turn about the quality of recommendations made is used to train the agent and to improve future recommendations made by the system 100.
[024] In an embodiment, the system 100 itself may handle loading of the containers to appropriate slots, based on the loading sequence and the recommendations generated, using appropriate hardware means required for picking and placing the containers in the corresponding slots, thereby automating the container loading process. In another embodiment, the container loading sequence and the recommendations may be used as input by any external system, so as to perform the container loading.
[025] FIG. 3 is a flow diagram depicting steps involved in the process of rating container loading recommendations, using the system of FIG. 1, according to some embodiments of the present disclosure.
[026] Separately for each container loading recommendation generated or collectively for each container loading sequence generated, the system 100 identifies (302) corresponding shuffle count value. For instance, assume that for slot ‘A’, there are four matching containers ‘1’, ‘2’, ‘3’, and ‘4’. Further assume that there are 3 containers on top of (or beside, or in any other manner that blocks direct pick up of the container ‘1’ for loading) container ‘1’, two on container ‘2’, one on container ‘3’, and the container ‘4’ is placed on top of the stack. In this example, picking container 1 results in the shuffle count value being 3, picking container 2 results in the shuffle count value being 2, picking container 3 results in the shuffle count value being 1, and picking container 4 results in the shuffle count value being 0.
[027] The identified shuffle count value is then compared (304) with a threshold value of shuffle count. If the identified threshold value is less than the threshold value for a container loading recommendation or for a container loading sequence, then a positive rating is given to the container loading recommendation/ container loading sequence. Similarly, if the identified threshold value is exceeding the threshold value for a container loading recommendation, then a negative rating is given to the container loading recommendation/ container loading sequence. In an embodiment, the rating is provided separately for each container loading recommendation generated, wherein shuffle count value for each recommendation is compared with the threshold value. In another embodiment, the rating is provided collectively for each loading sequence generated, wherein a total shuffle count for a loading sequence is identified/estimated, and is compared with the threshold value.
[028] It is to be noted that though FIG. 3 is explained citing a simple example scenario involving 4 containers, the same process can be used for generating container loading sequence and recommendation for any number of slots and containers.
[029] FIG. 4 is an example state space representation of container and slot data, according to some embodiments of the present disclosure. By virtue of the state-space representation, the system 100 displays slots and corresponding matching containers. Here, the slots are represented in ‘storage space state’, and containers to be filled in the slots are represented in ‘yard state’. For each slot in the storage space state, matching containers in the yard state are highlighted (refer to the blocks in the storage space state and the yard state that are highlighted in rectangle blocks). As mask-ids of the containers and the slots are highlighted in the state-space representation.
[030] The written description describes the subject matter herein to enable any person skilled in the art to make and use the embodiments. The scope of the subject matter embodiments is defined by the claims and may include other modifications that occur to those skilled in the art. Such other modifications are intended to be within the scope of the claims if they have similar elements that do not differ from the literal language of the claims or if they include equivalent elements with insubstantial differences from the literal language of the claims.
[031] The embodiments of present disclosure herein addresses unresolved problem of container loading sequence generation. The embodiment, thus provides a mechanism for generating a container loading sequence based on container loading recommendations. Moreover, the embodiments herein further provides mechanism to train the system based on ratings generated for each container loading recommendation generated and based on other relevant data, based on a Reinforcement Learning (RL) mechanism, so as to improve (future) recommendations/recommendations generated at a later point of time .
[032] It is to be understood that the scope of the protection is extended to such a program and in addition to a computer-readable means having a message therein; such computer-readable storage means contain program-code means for implementation of one or more steps of the method, when the program runs on a server or mobile device or any suitable programmable device. The hardware device can be any kind of device, which can be programmed including e.g. any kind of computer like a server or a personal computer, or the like, or any combination thereof. The device may also include means which could be e.g. hardware means like e.g. an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or a combination of hardware and software means, e.g. an ASIC and an FPGA, or at least one microprocessor and at least one memory with software modules located therein. Thus, the means can include both hardware means and software means. The method embodiments described herein could be implemented in hardware and software. The device may also include software means. Alternatively, the embodiments may be implemented on different hardware devices, e.g. using a plurality of CPUs.
[033] The embodiments herein can comprise hardware and software elements. The embodiments that are implemented in software include but are not limited to, firmware, resident software, microcode, etc. The functions performed by various modules described herein may be implemented in other modules or combinations of other modules. For the purposes of this description, a computer-usable or computer readable medium can be any apparatus that can comprise, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device.
[034] The illustrated steps are set out to explain the exemplary embodiments shown, and it should be anticipated that ongoing technological development will change the manner in which particular functions are performed. These examples are presented herein for purposes of illustration, and not limitation. Further, the boundaries of the functional building blocks have been arbitrarily defined herein for the convenience of the description. Alternative boundaries can be defined so long as the specified functions and relationships thereof are appropriately performed. Alternatives (including equivalents, extensions, variations, deviations, etc., of those described herein) will be apparent to persons skilled in the relevant art(s) based on the teachings contained herein. Such alternatives fall within the scope and spirit of the disclosed embodiments. Also, the words “comprising,” “having,” “containing,” and “including,” and other similar forms are intended to be equivalent in meaning and be open ended in that an item or items following any one of these words is not meant to be an exhaustive listing of such item or items, or meant to be limited to only the listed item or items. It must also be noted that as used herein and in the appended claims, the singular forms “a,” “an,” and “the” include plural references unless the context clearly dictates otherwise.
[035] Furthermore, one or more computer-readable storage media may be utilized in implementing embodiments consistent with the present disclosure. A computer-readable storage medium refers to any type of physical memory on which information or data readable by a processor may be stored. Thus, a computer-readable storage medium may store instructions for execution by one or more processors, including instructions for causing the processor(s) to perform steps or stages consistent with the embodiments described herein. The term “computer-readable medium” should be understood to include tangible items and exclude carrier waves and transient signals, i.e., be non-transitory. Examples include random access memory (RAM), read-only memory (ROM), volatile memory, nonvolatile memory, hard drives, CD ROMs, DVDs, flash drives, disks, and any other known physical storage media.
[036] It is intended that the disclosure and examples be considered as exemplary only, with a true scope and spirit of disclosed embodiments being indicated by the following claims.
,CLAIMS:1. A processor implemented method (200) for generating recommendations for container loading, comprising:
identifying (202) a plurality of containers (Ci) corresponding to each of a plurality of slots (Si), via one or more hardware processors;
generating (204) a plurality of container loading recommendations based on the plurality of containers identified for each of the plurality of slots, via the one or more hardware processors;
determining (206) a shuffle count value for each of the plurality of container loading recommendations, via the one or more hardware processors, wherein the shuffle count value of a container-slot pair in the plurality of container loading recommendations represents number of other containers to be moved so as to pick the container to load to the slot, at a time instance ti;
providing (208) one of a positive rating or a negative rating for each of the plurality of container loading recommendations, based on corresponding shuffle count value, via the one or more hardware processors; and
picking (210) one container each as a target container for each of the plurality of slots, via the one or more hardware processors, wherein the target container has least value of the shuffle count among all the containers having the positive rating from the plurality of containers corresponding to each slot.
2. The method as claimed in claim 1, wherein identifying the plurality of containers corresponding to each of the plurality of slots comprises:
comparing a mask-id of each of the plurality of containers with mask-id of each of the plurality of slots; and
identifying all containers having the mask-id same as that of a slot as matching containers corresponding to the slot.
3. The method as claimed in claim 1, wherein providing one of the positive rating or the negative rating comprises:
comparing (304) the shuffle count value corresponding to each of the plurality of container loading recommendations with a threshold of shuffle count;
assigning (308) the positive rating if the shuffle count value is less than the threshold of shuffle count; and
assigning (310) the negative rating if the shuffle count value exceeds the threshold of shuffle count.
4. A system (100) for generating recommendations for container loading, comprising:
a memory module (101) storing a plurality of instructions;
one or more communication interfaces (103); and
one or more hardware processors (102) coupled to the memory module (101) via the one or more communication interfaces (103), wherein the one or more hardware processors are caused by the plurality of instructions to:
identify (202) a plurality of containers (Ci), corresponding to each of a plurality of slots (Si);
generate (204) a plurality of container loading recommendations based on the plurality of containers identified for each of the plurality of slots;
determine (206) a shuffle count value for each of the plurality of container loading recommendations, wherein the shuffle count value of a container-slot pair in the plurality of container loading recommendations represents number of other containers to be moved so as to pick the container to load to the slot, at a time instance ti;
provide (208) one of a positive rating or a negative rating for each of the plurality of container loading recommendations, based on corresponding shuffle count value; and
pick (210) one container each, as a target container for each of the plurality of slots, wherein the target container has least value of the shuffle count among all the containers having the positive rating from the plurality of containers corresponding to each slot.
5. The system (100) as claimed in claim 4, wherein the system (100) identifies the plurality of containers corresponding to each of the plurality of slots by:
comparing a mask-id of each of the plurality of containers with mask-id of each of the plurality of slots; and
identifying all containers having the mask-id same as that of a slot as matching containers corresponding to the slot.
6. The system (100) as claimed in claim 4, wherein the system (100) provides one of the positive rating or the negative rating by:
comparing (304) the shuffle count value corresponding to each of the plurality of container loading recommendations with a threshold of shuffle count;
assigning (308) the positive rating if the shuffle count value is less than the threshold of shuffle count; and
assigning (310) the negative rating if the shuffle count value exceeds the threshold of shuffle count.