WO2022264117A1

WO2022264117A1 - Dynamic buffer management for inventory control

Info

Publication number: WO2022264117A1
Application number: PCT/IL2021/050738
Authority: WO
Inventors: Motoi Tobita
Original assignee: Onebeat Ltd.
Priority date: 2021-06-17
Filing date: 2021-06-17
Publication date: 2022-12-22

Abstract

A dynamic buffer management (DBM) system for inventory control communicates with inventory locations. Each of the inventory locations stocks multiple stock keeping units (SKUs). For each of the SKUs at each of the locations the DBM system: maintains buffer data for the SKU, the buffer data including a respective buffer size for the SKU at the location; estimates inventory levels based on data input over a data interface; calculates respective ranges of zones within the buffer size based on specified ratios for the estimated inventory levels being within each of the zones; and outputs a trigger for recalculating the buffer size when a presence of historical inventory levels within at least one of the calculated ranges exceeds a respective threshold for the zone and maintains the buffer size unchanged when the presence of the historical inventory levels is within the respective threshold.

Description

DYNAMIC BUFFER MANAGEMENT FOR INVENTORY CONTROL

FIELD AND BACKGROUND OF THE INVENTION

The present invention, in some embodiments thereof, relates to a dynamic buffer management (DBM) system and, more particularly, but not exclusively, to a DBM system with automatic parameter calculation.

Inventory control is the process of managing stock once it arrives at a location, such as a store or storage facility. Inventory control attempts to optimize the correct amount of inventory for each stock keeping unit (SKU) at each location. An SKU is a distinct type of object which is maintained in inventory. For example, an SKU may be a shirt of a particular design, in a specific size and color. Even a small store is likely to have a large number of SKUs, each needing inventory control. The number of SKUs needing inventory control increases greatly in larger stores or chains of stores. This task is difficult for an individual to perform without specialized tools, especially since inventory levels must be frequently monitored and updated in order to take into account possible spikes in demand and to prevent overstocking.

Inventory control systems are analytical and technological tools for inventory control. Inventory control systems aim to actualize the situation to have the right item, in the right amount and at the right time.

The current standard of inventory management in various industries is the “Min-MAX method.” Any inventory control system that implements the Min-MAX method needs configuration parameters to control inventory. To determine the configuration parameters, “average daily consumption” plays a significant role. More sophisticated versions of existing ERP systems use the average daily consumption, combined with other factors such as the standard deviation of the consumption, shelf size, order lot size, etc. to represent their business environment more closely.

Another inventory control system is Demand Driven Material Requirements Planning (DDMRP). DDMRP uses the average daily consumption to determine the configuration parameters.

An additional inventory control system is dynamic buffer management (DBM). DBM systems define a buffer for each SKU at each location. The buffer has three zones which define inventory levels that are considered to be in an oversupply zone (i.e. high inventory level), a target zone (i.e. suitable amount of inventory) and an undersupply zone (i.e. shortage of inventory). When the actual inventory is in the oversupply zone (also denoted herein the green zone) for too long a time period, the buffer size is decreased. When the actual inventory is in the undersupply zone (also denoted herein the red zone) for too long a time period, the buffer size is increased. In this way the inventory for the SKU at the given location is maintained within the target zone (also denoted herein the yellow zone) most of the time.

In current DBM systems, critical parameters for the system are set manually. Even when the scale of inventory management is moderate, a skilled person in charge of managing inventory will have difficulty finding an optimum parameter set for each SKU at each location. As a result, those parameters are left unmaintained, leading to shortages and surpluses.

Additional background art includes:

[1] K. Dalai and S. Shah, “Dynamic Buffer Management,” International Journal of Innovative Research in Advanced Engineering (IJIRAE) ISSN: 2349-2163 Issue 7, Volume 2 (July 2015).

SUMMARY OF THE INVENTION

It is an object of the present disclosure to describe a system, a method and a computer program product for a DBM system with automatic parameter calculation.

In embodiments of the invention, a DBM system maintains buffer data for each SKU at each location. Buffer data, including the respective buffer size, is maintained by the DBM system for each SKU at each location. The buffer size and other DBM system parameters are used to control inventory for each SKU at each location. As described in more detail below, embodiments of the invention use a “significance test” to automatically control buffer size with great accuracy, thereby avoiding the sub-optimal average daily consumption model of current systems.

Typically the buffer is divided into at least three zones: a) The oversupply zone (denoted the green zone) - Indicates a high inventory level. b) The target zone (denoted the yellow zone) - A suitable amount of inventory; c) The undersupply zone (denoted the red zone) - shortage of inventory.

In situations where the buffer size is very small (e.g. one or two), the buffer may be divided into two zones, as described in more detail below.

One or more parameters for managing inventory for an SKU at a location (also denoted an SL-SKU) are calculated automatically. These parameters may include but are not limited to: a) Parameter 1 - The range of the red zone within the buffer size (specified for example as a percentage); b) Parameter 2 - The range of the green zone within the buffer size (specified for example as a percentage); c) Parameter 3 - A threshold for buffer size increase; d) Parameter 4 - A threshold for buffer size decrease; e) Parameter 5 - A buffer size increase (specified for example as a percentage); f) Parameter 6 - A buffer size decrease (specified for example as a percentage).

Inventory levels for each SL-SKU are estimated by the DBM system based on historical data provided by the respective location (e.g. consumption data for the SKU in recent days). The estimated inventory levels for the SL-SKU are analyzed in order to calculate DBM system parameters, as described in more detail below.

By automatically calculating the respective buffer parameters for each SL-SKU, embodiments of the DBM system and method described herein provide the benefit of dramatically reducing the burden on the person(s) in charge of managing inventory. An additional benefit is that the DBM system optimizes the respective parameters for each SL-SKU, resulting in much more effective inventory control than when using the rule of thumb values that are typically provided as user input in current DBM systems. Yet another benefit is that because the parameters are calculated automatically, inventory control parameters may be found for tens of millions of items within a realistic amount of time.

As a result of better inventory control, the user of the system may reduce shortages without having more inventory and/or reduce the inventory levels without jeopardizing service level.

When shortage is reduced, a natural expectation is higher sales. Because shortages usually occur for good-selling SL-SKUs (relative to average- selling items), reducing 1% of shortages, for example, often contributes to a few percent increase in sales. When excess inventory is reduced, the discounts and disposal are also reduced. For example, when there is excess inventory at retail stores, store owners try to sell out those items by setting discount prices. Sometimes the discount levels are very deep, which may damage the financial performance of the company. When there is leftover inventory regardless of these efforts, the leftover inventory is disposed of. Disposal wastes money (and possibly other resources) because the item has been invested in from securing raw material to the final product, shipment etc. So, reducing the discount and disposal at the root cause (in other words at the ordering level) directly contributes to reducing unnecessary costs.

Additionally, being able to operate with lower level of inventory benefits product quality. If there is quality problem, it is easier to track it and make corrections when the inventory level is lower. Financially, maintaining a lower inventory without worsening the shortage ration results in the company turning investment into sales faster. Usually, the company measures the performance of inventory management by “inventory turns”. Embodiments of the instant invention greatly improves the inventory turns, which results in a better cash flow for the company.

According to first aspect of some embodiments of the present invention there is provided a dynamic buffer management (DBM) system for inventory control. The DBM system includes: a processing circuitry and a data interface configured for communicating over a network with multiple inventory locations, each of the inventory locations stocking multiple stock keeping units. The processing circuitry is configured to: for each of the stock keeping units at each of the locations: maintain buffer data for the stock keeping unit (SKU) at the location, the buffer data comprising a respective buffer size for the SKU; estimate inventory levels based on data input over the data interface; calculate respective ranges of zones within the buffer size based on specified ratios for the estimated inventory levels being within each of the zones; and output a trigger for recalculating the buffer size when a presence of historical inventory levels within at least one of the calculated ranges exceeds a respective threshold for the zone and maintain the buffer size unchanged when the presence of the historical inventory levels is within the respective threshold.

According to second aspect of some embodiments of the present invention there is provided a method for inventory control. The method includes: inputting, from inventory locations stocking multiple stock keeping units, respective data for estimating inventory levels of the stock keeping units at the locations over time; for each of the stock keeping units at each of the locations: maintaining buffer data for the stock keeping unit (SKU) at the location, the buffer data comprising a respective buffer size for the SKU at the location; estimating inventory levels of the stock keeping unit at the location using the respective input data for the SKU at the location; calculating respective ranges of multiple zones within the buffer size based on specified ratios for the estimated inventory levels being within each of the zones; and outputting a trigger for recalculating the buffer size when a presence of historical inventory levels within at least one of the calculated ranges exceeds a specified threshold and maintaining the buffer size unchanged when the presence of the historical inventory levels within the at least one of the recalculated ranges is within the specified threshold.

According to some embodiments of the invention, the data for estimating the inventory levels comprises consumption data for the stock keeping unit at the location.

According to some embodiments of the invention, the buffer size is recalculated when the trigger is output and the recalculated buffer size is output to the location over the data interface.

According to some embodiments of the invention, the zones include a first zone indicative of a high level of inventory, a second zone indicative of a target amount of inventory and a third zone indicative of an undersupply of inventory.

According to some embodiments of the invention, a buffer increase trigger is output when a time period the historical inventory levels are present in the third zone exceeds a first threshold, the first threshold being calculated from a probability of the estimated inventory levels remaining in the third zone for consecutive time periods.

According to some embodiments of the invention, a buffer increase trigger is output when a penetration of the historical inventory levels within the third zone exceeds a second threshold, the first penetration level being calculated from a probability of the estimated inventory levels remaining at each level of the third zone for consecutive time periods.

According to some embodiments of the invention, a buffer decrease trigger is output when a time period the historical inventory levels are present in the first zone exceeds a third threshold, the third threshold being based on a probability of the estimated inventory levels remaining in the first zone for consecutive time periods.

According to some embodiments of the invention, a buffer decrease trigger is output when a penetration of the historical inventory levels within the first zone exceeds a fourth threshold, the fourth threshold being based on a probability of the estimated inventory levels remaining at each of level of the first zone for consecutive time periods.

According to some embodiments of the invention, the buffer size is decreased when the historical inventory levels remain outside the third zone longer than a first specified time period.

According to some embodiments of the invention, the buffer size is decreased when the historical inventory levels increase for longer than a second specified time period.

According to some embodiments of the invention, the buffer size is increased when the historical inventory levels decrease for longer than a third specified time period.

According to some embodiments of the invention, a multiple for buffer size increase is based on the recalculated ranges of the second zone and the third zone. According to some embodiments of the invention, a multiple for buffer size decrease is based on the recalculated ranges of the first zone and the second zone.

According to some embodiments of the invention, an indicator of anomalies is output to the location over the data interface when anomalies are identified in the estimated inventory levels.

According to some embodiments of the method according to the second aspect of the invention, a computer program comprising program code stored on a computer readable medium carries out the method when executed by one or more processors.

The foregoing and other objects are achieved by the features of the independent claims. Further implementation forms are apparent from the dependent claims, the description and the figures.

Other systems, methods, features, and advantages of the present disclosure will be or become apparent to one with skill in the art upon examination of the following drawings and detailed description. It is intended that all such additional systems, methods, features, and advantages be included within this description, be within the scope of the present disclosure, and be protected by the accompanying claims.

Unless otherwise defined, all technical and/or scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the invention pertains. Although methods and materials similar or equivalent to those described herein can be used in the practice or testing of embodiments of the invention, exemplary methods and/or materials are described below. In case of conflict, the patent specification, including definitions, will control. In addition, the materials, methods, and examples are illustrative only and are not intended to be necessarily limiting.

Implementation of the method and/or system of embodiments of the invention can involve performing or completing selected tasks manually, automatically, or a combination thereof. Moreover, according to actual instrumentation and equipment of embodiments of the method and/or system of the invention, several selected tasks could be implemented by hardware, by software or by firmware or by a combination thereof using an operating system.

For example, hardware for performing selected tasks according to embodiments of the invention could be implemented as a chip or a circuit. As software, selected tasks according to embodiments of the invention could be implemented as a plurality of software instructions being executed by a computer using any suitable operating system. In an exemplary embodiment of the invention, one or more tasks according to exemplary embodiments of method and/or system as described herein are performed by a data processor, such as a computing platform for executing a plurality of instructions. Optionally, the data processor includes a volatile memory for storing instructions and/or data and/or a non-volatile storage, for example, a magnetic hard-disk and/or removable media, for storing instructions and/or data. Optionally, a network connection is provided as well. A display and/or a user input device such as a keyboard or mouse are optionally provided as well.

BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWINGS

Some embodiments of the invention are herein described, by way of example only, with reference to the accompanying drawings. With specific reference now to the drawings in detail, it is stressed that the particulars shown are by way of example and for purposes of illustrative discussion of embodiments of the invention. In this regard, the description taken with the drawings makes apparent to those skilled in the art how embodiments of the invention may be practiced.

In the drawings:

FIG. 1 is a simplified schematic illustration of an exemplary DBM system, according to embodiments of the invention;

FIG. 2 is a simplified block diagram of a dynamic buffer management system according to embodiments of the invention;

FIGS. 3A-3D are simplified schematic illustrations of an exemplary buffer structure and exemplary parameters used by the DBM system;

FIG. 4 is a simplified flowchart of a method for triggering a buffer size change for a single SKU at a particular location, according to embodiments of the invention;

FIG. 5 illustrates an example of estimating inventory levels from previous inventory levels, consumption data and replenishment lead time;

FIG. 6A is an example of estimated inventory levels showing the transitions between black, red and yellow zones;

FIG. 6B is an example of monitoring accumulated penetration into the red zone;

FIG. 7A is a simplified flowchart of a method for updating a buffer size for a single SKU at a location, according to a first exemplary embodiment of the invention;

FIG. 7B is a simplified flowchart of a method for updating a buffer size for a single SKU at a location, according to a second exemplary embodiments of the invention;

FIG. 8 is a simplified schematic diagram of calculating parameters for a DBM system, according to an exemplary embodiment of the invention; and FIG. 9 is a simplified flowchart of a method for calculating a recommended new buffer size, according to an exemplary embodiment of the invention.

DESCRIPTION OF SPECIFIC EMBODIMENTS OF THE INVENTION

In embodiments of the invention, buffer data is maintained for each SL-SKU. The buffer data includes the respective buffer size and optionally additional data. The buffer is divided into at least three zones: a) The oversupply zone (denoted the green zone) - High inventory level. b) The target zone (denoted the yellow zone) - Suitable amount of inventory; c) The undersupply zone (denoted the red zone) - Shortage of inventory.

Optionally, the buffer also includes a black zone and/or a cyan zone. The black zone indicates that an SKU is out of stock (inventory=0). Optionally, when back orders are allowed (such as in E-commerce) the black zone also includes negative inventory levels. The cyan zone indicates that there is an excess of inventory.

At least some of the parameters used by the DBM system in order to analyze and control inventory levels are calculated automatically for each SL-SKU. These parameters may include but are not limited to:

1) The ranges of the red, green and yellow zones;

2) Buffer increase threshold;

3) Buffer decrease threshold;

4) Buffer increase percentage; and

5) Buffer decrease percentage.

The fluctuation of inventory levels are estimated from data, such as consumption data, provided by the locations. The estimated inventory levels analyzed for each SL-SKU based on its respective parameters. A trigger is output when the analysis for the SL-SKU determines that the buffer size should be recalculated.

Before explaining at least one embodiment of the invention in detail, it is to be understood that the invention is not necessarily limited in its application to the details of construction and the arrangement of the components and/or methods set forth in the following description and/or illustrated in the drawings and/or the Examples. The invention is capable of other embodiments or of being practiced or carried out in various ways. Embodiments may be a system, a method, and/or a computer program product. The computer program product may include a computer readable storage medium (or media) having computer readable program instructions thereon for causing a processor to carry out aspects of the embodiments.

The computer readable storage medium can be a tangible device that can retain and store instructions for use by an instruction execution device. The computer readable storage medium may be, for example, but is not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. A non-exhaustive list of more specific examples of the computer readable storage medium includes the following: a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, and any suitable combination of the foregoing. A computer readable storage medium, as used herein, is not to be construed as being transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through a waveguide or other transmission media (e.g., light pulses passing through a fiber-optic cable), or electrical signals transmitted through a wire.

Computer readable program instructions described herein can be downloaded to respective computing/processing devices from a computer readable storage medium or to an external computer or external storage device via a network, for example, the Internet, a local area network, a wide area network and/or a wireless network. The network may comprise copper transmission cables, optical transmission fibers, wireless transmission, routers, firewalls, switches, gateway computers and/or edge servers. A network adapter card or network interface in each computing/processing device receives computer readable program instructions from the network and forwards the computer readable program instructions for storage in a computer readable storage medium within the respective computing/processing device.

Computer readable program instructions for carrying out operations of embodiments may be assembler instructions, instruction-set-architecture (ISA) instructions, machine instructions, machine dependent instructions, microcode, firmware instructions, state-setting data, or either source code or object code written in any combination of one or more programming languages, including an object oriented programming language such as Smalltalk, C++ or the like, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The computer readable program instructions may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet Service Provider). In some embodiments, electronic circuitry including, for example, programmable logic circuitry, field-programmable gate arrays (FPGA), or programmable logic arrays (PLA) may execute the computer readable program instructions by utilizing state information of the computer readable program instructions to personalize the electronic circuitry, in order to perform aspects of embodiments.

Aspects of embodiments are described herein with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer readable program instructions.

These computer readable program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks. These computer readable program instructions may also be stored in a computer readable storage medium that can direct a computer, a programmable data processing apparatus, and/or other devices to function in a particular manner, such that the computer readable storage medium having instructions stored therein comprises an article of manufacture including instructions which implement aspects of the function/act specified in the flowchart and/or block diagram block or blocks.

The computer readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable apparatus or other device to produce a computer implemented process, such that the instructions which execute on the computer, other programmable apparatus, or other device implement the functions/acts specified in the flowchart and/or block diagram block or blocks.

The flowchart and block diagrams in the Figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems that perform the specified functions or acts or carry out combinations of special purpose hardware and computer instructions.

The DBM System

Referring now to the drawings, Fig. 1 is a simplified schematic illustration of an exemplary DBM system, according to embodiments of the invention. Fig. 1 illustrates embodiments in which the DBM system operates as a back end, gathering and analyzing information from multiple locations. This embodiment may be suitable for a chain store which maintains inventory centrally for many locations.

DBM system 100 communicates over network 110 with a plurality of locations 120.1- 120. n. Each location stocks inventory for multiple SKUs, indicated 130.1-130.n respectively. DBM system 100 maintains respective buffer data for each SKU in inventories 130.1-130.n. The buffer data includes the buffer size for the SL-SKU and optionally respective ranges of zones within the buffer size. It is noted that the inventory may differ between locations (e.g., the types of SKUs maintained at the location and/or the respective inventory levels for each SKU).

In alternate embodiments, the DBM system is a standalone system which manages inventory levels for SKUs at a single location. A standalone embodiment may be suitable for a location which is not related to other locations but maintains inventory for a large number of SKUs (e.g. a hardware store which is not part of a chain of stores).

Optionally, DBM system 100 may be accessed by one or more Graphical User Interfaces (GUI). The locations of the GUIs may include but are not limited to:

1) Directly connected to or part of the DBM system (e.g. GUI 140);

2) At a location stocking inventory (e.g. GUI 150); and

3) Standalone GUI (e.g. GUI 160). As used herein the terms “stock keeping unit” and “SKU” mean a type of item or group of items whose inventory is managed using a single buffer.

As used herein the term “location” means a place (such as a store or warehouse) where one or more SKUs are stocked and/or consumed. It is noted that the device that communicates with the DBM system with regards to the inventory at a given location is not necessarily physically located at the place that the inventory is maintained.

As used herein the terms “stock keeping unit at a location” and “SL-SKU” mean a specific SKU at a specific location.

As used herein the term “inventory level” means the amount of the SKU available as inventory for consumption. The units in which the inventory level is specified may be based on the type of SKU (e.g. as a number, volume, weight, etc.).

As used herein the term “zone” means a range of inventory levels within the total buffer size.

As used herein the term “buffer size” means the maximum amount of inventory that should be available for the SKU at the location.

As used herein the term “estimated inventory levels” means estimates of respective inventory levels for a sequence of preceding time periods. For example, when the inventory level is monitored daily and the estimated inventory levels are of length fifty, the “estimated inventory levels” is a sequence of inventory levels estimated over the previous fifty days.

As used herein the term “historical inventory levels” means the actual inventory levels of the SKU at the location for a sequence of time periods. The actual inventory level of an SL-SKU for a given time period is the inventory level at the location at the end of the time period. For example, when the inventory level is monitored daily and the historical inventory levels are of length fifty, the “historical inventory levels” is a sequence of inventory levels at the end of each of the previous fifty days.

The estimated inventory levels are used to determine the DBM system parameters. The historical inventory levels are analyzed to determine the need for a buffer size increase or decrease based on the DBM system parameter values.

Optionally the inventory level is the amount of inventory available for consumption at the location. The inventory level includes the amount of inventory stocked at the location and optionally one or both of: a) Inventory in transit to the location; and b) Inventory in production that is to be replenished to the location. Optionally, an order is placed when the inventory level is lower than the buffer size. Further optionally, the order amount is the difference between the inventory level and the buffer size. As noted above, the inventory level may include inventory in transit and/or in production in addition to the inventory at the location.

For example, when the buffer size does not change and orders are placed daily, the amount ordered in a given day equals the amount consumed that day. When the buffer size is increased, the amount ordered in a given day equals the amount consumed that day plus the increase in the buffer size on the previous day.

Alternately or additionally, the order amount is constrained by other factors which are taken into consideration when placing the order. These factors may include but are not limited to: a) The minimum order size - When a minimum order size is specified for the SL-SKU, an order is not placed until the difference between the inventory level and the buffer size is at least the minimum order size. b) Last batch replenishment - The last batch replenishment is a parameter that is between 0 and 1 that may be used when the order batch size is more than 1. For example when the order amount is 10 and the last batch replenishment value is 0.3, an order will be placed when inventory level is 3 units (10 times 0.3) less than the buffer size.

Reference is now made to Fig. 2, which is a simplified block diagram of a dynamic buffer management system according to embodiments of the invention. DBM system 200 includes processing circuitry 210 which communicates with the locations via communication interface 220. Optionally, DBM system 200 includes internal memory 230 which stores information such as respective buffer data for one or more of the SL-SKUs.

Reference is now made to Figs. 3A-3D, which are simplified schematic illustrations of an exemplary buffer structure and exemplary parameters used by the DBM system.

Fig. 3A shows a buffer which includes three zones, green zone 310, yellow zone 320 and red zone 330. The respective ranges of the zones are set so that the estimated inventory levels over time are split amongst the zones with a specified ratio. Optionally, this ratio is 16%:68%:16%. An example is described below (for Fig. 5).

Fig. 3B is a simplified illustration of thresholds within the zones. A buffer decrease is triggered when the threshold is crossed in green zone 310. A buffer increase is triggered when the threshold is crossed in red zone 330. According to embodiments of the invention, these thresholds are calculated from the estimated inventory levels and a significance level, typically set by the user. As a result, a significance test may be performed to detect a trigger for buffer size increase or decrease. Fig. 3C illustrates buffer size increase. When a buffer size increase is triggered, the buffer size may be increased by a calculated or fixed increase percentage.

Fig. 3D illustrates buffer size decrease. When a buffer size decrease is triggered, the buffer size may be decreased by a calculated or fixed decrease percentage.

Referring again to Fig. 1, each location 120.1-120.n provides DBM system 100 with the data needed by DBM 100 in order to estimate inventory levels for each SKU at the respective location. DBM system 100 analyzes this data (along with other parameters as described below) in order to trigger a buffer size increase or decrease as needed for each SL-SKU.

Optionally, the location provides DBM system 100 with respective consumption data for at least one SKU.

As used herein, the term “consumption data” means the quantity of the SKU that was removed from the inventory during the preceding time period.

In one example, if the SKU is a shirt and three of these shirts were purchased today, the consumption data is three. In a second example, the SKU is sold by weight and the buffer size is checked weekly; in this case the consumption data is the weight of the SKU that was purchased in the previous week.

Optionally, when a buffer size recalculation is triggered DBM system 100 recalculates a buffer size based on parameters stored by DBM 100. DBM system 100 provides the updated buffer size to the relevant location. Alternately or additionally, when a buffer size recalculation is triggered, DBM system 100 forwards the trigger to the location along with an indication of whether the buffer should be increased or decreased. When a buffer size recalculation is not triggered, the buffer size is maintained at its current level.

When the buffer size is recalculated, the new buffer size is updated in the respective buffer data for use in the next iteration.

Optionally, DBM system 100 calculates at least one parameter which is used to control the buffer size. These parameters may include but are not limited to:

1) The ranges of the red, green and yellow zones;

2) Buffer increase threshold;

3) Buffer decrease threshold;

4) Buffer increase percentage; and

5) Buffer decrease percentage.

Embodiments for calculating these parameters are presented below.

The frequency at which inventory level analysis is performed typically depends on the client’s business environment. Analysis of the historical inventory levels is typically performed every day. However, not all SKUs needs buffer size adjustment every time the analysis is performed. The buffer size may be adjusted only for SKUs having historical inventory levels which exceed the specified threshold. Typically, in highly active environments (e.g. grocery stores) approximately 10% of the SKUs need buffer size adjustment at least once in a given day. In slow moving environments (e.g. apparel stores), the buffer adjustment is typically less frequent. When the DBM system 100 analyzes the historical inventory levels daily, the buffer size may be recalculated and updated without delay.

Modes of operation of DBM system 100 include but are not limited to: a) Full- automation mode - The buffer size is updated automatically by the DBM system when a trigger is output and the new size is provided to the location. b) Semi- automation mode - In this mode the DBM system analyzes the consumption data in order to detect abnormal consumption of an SKU. When abnormal behavior is detected, the system alerts the client to check this particular SKU manually. Normal change may be handled by DBM system 100 and the buffer size may be adjusted automatically. c) Manual mode - In this mode the DBM system outputs the trigger to the end user. The end user decides whether to change the buffer size.

Optionally, different SL-SKUs operate in different modes.

Optionally, DBM system 100 operates in two phases. In the initial phase, DBM system 100 calculates parameters (e.g. zones and thresholds) for multiple SL-SKUs. The respective buffer sizes are recalculated for SL-SKUs for which a trigger was detected during a subsequent execution phase. Optionally, some or all of the parameters are recalculated during the execution phase.

The location, operating mode and frequency of buffer update may be the same or different for multiple SKUs and/or locations. This approach has the benefit of giving the client flexibility in how inventory is controlled, based on the requirements and constraints of particular SL-SKUs and/or the organization.

Reference is now made to Fig. 4, which is a simplified flowchart of a method for triggering a buffer size change for a single SKU at a particular location, according to embodiments of the invention.

Buffer data is maintained for each SL-SKU. The buffer data includes the respective buffer size. Optionally, the buffer data includes additional information. This information may include but is not limited to:

1) Estimated inventory levels for the SL-SKU; 2) Respective ranges of the zones within the buffer;

3) Buffer increase and/or decrease threshold(s);

4) Buffer increase and/or decrease percentages; and

5) User set variables such as significance, lead time, target shortage, minimum buffer size, last batch replenishment, etc.

In 410, data for estimating inventory levels of the stock keeping unit over time is input from the location. Optionally, this data is the consumption of the SKU at the location. For example, at the end of every day the location may automatically send the DBM system the amount of the SKU that was consumed that day.

In 420, the inventory levels for the SL-SKU are estimated using the data provided by the location and, optionally, the previous estimated inventory levels.

Optionally, the estimated inventory levels include data for a fixed number of time periods. For example, if the estimated inventory levels include estimated inventory levels for fifty days, the inventory level from 50 days earlier is dropped when a new inventory level is estimated.

Optionally, data for inventory level estimation is also obtained from other sources. For example, at startup the initial estimated inventory levels may be based on historical data from a similar SL-SKU.

Optionally, the initial inventory level at startup is the maximum expected consumption during the replenishment lead time.

In 430, the thresholds are determined and the ranges of the zones within the current buffer size are calculated based on the estimated inventory levels.

The threshold(s) may be calculated according to any of the embodiments described herein. Optionally, one or both of the thresholds are calculated using a significance test, as described in more detail below. Further optionally, the thresholds are set so that a buffer size increase calculation is triggered when the historical inventory levels have a significant penetration into the red zone and a buffer size decrease calculation is triggered when the historical inventory levels have a significant penetration into green zone.

Considering each of the estimated inventory level as a data point, the ranges are calculated so that the number of data points within each of the zones is the same as or close to a specified ratio. The relative sizes of each of the zones within the buffer size will not necessarily be the same as the specified ratio used to establish the ranges themselves. The ranges of the zones are based on an analysis of the estimated inventory levels over time. The ratio of the zones within the buffer size reflects the relative sizes of the zones at a given point in time. Optionally, the zones in the buffer are defined so as to reduce the shortages as much as possible. In this case, the buffer size is the difference between the maximum and minimum levels in the estimated inventory levels. This buffer size is divided into green, yellow, and red zones so that the number of points in the estimated inventory levels that lie within each zone are as close as possible to a specified ratio, such as 16:68:16. In alternate embodiments, there is a shortage target, such as X%. If the specified ratio is 16:68:16, the difference between the maximum and minimum inventory levels in the estimated inventory levels is divided into four zones of 16:68:(16-X):X. The first three zones are the green yellow and red zones (16, 68 and 16-X).

In 440, a trigger for recalculating the buffer size is output when the presence of historical inventory levels within at least one of the recalculated ranges exceeds the respective threshold. Optionally, the trigger indicates whether the buffer size should be increased or decreased (e.g. the trigger signal is different for increase and decrease, the trigger includes a flag bit indicating increase or decrease, etc.). When the presence of the historical inventory levels is within the specified threshold, no trigger is output and therefore the buffer size is not changed.

Optionally, the method further includes recalculating the buffer size when a trigger is detected and outputting the recalculated buffer size to the location. The buffer size is recalculated based on the type of trigger output. When the trigger indicates that there is too much inventory (too much green), the buffer size is decreased by the buffer decrease percentage. When the trigger indicates that there is too little inventory (too much red), the buffer size is increased by the buffer increase percentage.

Typically the DBM system performs these steps for each SL-SKU independently. It is noted that although the basic approach is substantially the same for each SL-SKU, inventory may be managed differently for each SL-SKU, based on parameters such as lead time, minimum buffer size and/or the frequency at which the buffer size is checked and possibly recalculated.

Optionally, the method further includes identifying anomalies in the estimated inventory levels and outputting an indicator of the anomalies to the location. Such anomalies may be reduced by filtering the input data prior to estimating the inventory levels, as described in more detail below.

As used herein the term “presence of the historical inventory levels within a zone exceeds a threshold” means that a calculation based on the historical inventory levels resulted in a value that exceeds the threshold.

As will be appreciated by the skilled person, exceeding a threshold may mean being greater than or lesser than a specified value based on how the threshold is defined. It is therefore noted that the term “exceeds a threshold” does not require that the mathematical result of the analysis is greater than a specified value.

It is further noted that the term “exceeds a threshold” does not indicate whether the inventory level should be increased or decreased. The determination of whether the inventory level should be decreased or increased is based on whether the threshold that has been exceeded is the green zone threshold (i.e. TMG) or the red zone threshold (i.e. TMR).

Estimating inventory levels

Optionally, the inventory levels are estimated based on previous inventory levels, consumption data and lead time for inventory replenishment. In an exemplary embodiment, the inventory level for the most recent time period is calculated as:

Estimated inventory for the recent time period = Estimated inventory level from the preceding time period + the amount replenished in the recent time period - the amount consumed in the recent time period.

Table 1 is an example of estimating inventory levels according to the equation above. Starting with an inventory level of 10 on the first day, the estimated inventory levels decrease when an SKU is consumed (days 2, 7 and 8) and increase when the SKU is replenished (days 5 and 11). In Table 1, a replenishment order is placed every other day (days 2, 4, 6, 8 and 10). The item is replenished three days after the order is placed. Therefore the amount ordered on day 2 arrives on day 5. The amount ordered on day 8 (which is the sum of the consumption on days 7 and 8) arrives on day 10.

Table 1

Optionally, the total lead time (denoted herein the replenishment lead time) is based on several components. For example, the replenishment lead time may equal the order lead time plus production and/or transportation lead times.

As used herein the term “replenishment lead time” means the total amount of time between when a replenishment request is placed to the actual replenishment of the SKU in the inventory. As used herein the term “production lead time” means a portion of the replenishment lead time that is caused by the time required to produce the item after the replenishment request is made.

As used herein the term “transportation lead time” means a portion of the replenishment lead time that is caused by the time required to transport to the physical location where the item is stored after the replenishment request is made.

In Table 1 the replenishment lead time is four days. One day is the order lead time (because the amount consumed on odd numbered days is only ordered on the following day) and three days for transportation lead time. Reference is now made to Fig. 5 which illustrates an example of estimating inventory levels from previous inventory levels, consumption data and replenishment lead time. In Fig. 5, the replenishment lead time is five days. During the first five days the SKU is consumed but there is no replenishment, so the estimated inventory levels decline steadily. On day 6 the first replenishment occurs and the inventory level rises as a result of having replenished slightly more than the consumed amount. From this time forward, the replenishment data equals the consumption data with a time lag equal to the five day replenishment lead time. On day 12, for example, the inventory level from day 11 equals 11, the replenishment equals 15 and the consumption equals 13. The estimated inventory level on day 12 is therefore 11+15-13=13. Note that the replenishment equals the consumption from five days earlier on day 7.

Recalculating zones

The zones are recalculated after the estimated inventory levels have been updated. The zones are defined so that the estimated inventory levels are present in each of the zones as close as possible to a specified ratio.

For example, consider the case where the specified ratio for the green, yellow and red zones is 16%:68%:16% and the estimated inventory includes fifty data points. The goal is to specify the ranges so that 8 data points are in the green zone, 34 data points are in the yellow zone and 8 data points are in the red zone (or red and black zone).

For example, in Fig. 5 the maximum amount of inventory is 28. In order to have the desired number of data points in each zone, the green zone range is set to 22-28 and the yellow zone range is set to 13-22. The target shortage ratio is 2%, so one data point is in the black zone. Therefore, the red zone is set to 10-13. The entire buffer (without the black zone) ranges from 10 to 28. Therefore, the percentage of the green zone in the buffer is (28-22)/18=0.33; the percentage of the yellow zone in the buffer is (22- 13)/18=0.5; and the percentage of the red zone in the buffer is ( 13- 10)/18=0.17.

Attaining a division that is close to the specified ratio is relatively easy when the buffer size is large. However, in retail environments the buffer size is often less than ten. When the buffer size is small it may be difficult to divide the buffer into ranges which are close to the specified ratio. In this case, the buffer is optionally divided into ranges which are as close as possible to the desired ratio.

For example, consider Table 2 which represents estimated inventory levels for 33 days. The estimated inventory level was six during four of the 33 days, five during twelve of the 33 days, and so forth.

Table 2

The goal is for the red, yellow and green zones percentages to be as close as possible to 16%:68%:16%. If the green zone is set to 6, the percentage of days that the inventory level is in the green zone equals 12% (since 4/33=.12). If the green zone is set to 5-6, the percentage of days that the inventory level is in the green zone is 48% (since 16/33=.48). The ratio of 16% to 12% (i.e. 1.3) is less than the ratio of 48% to 16% (i.e. 3). Therefore the green zone is set to 6.

Similarly, if the red zone is defined as 1-2, the percentage of days that the inventory level is in the red zone equals 9% (since 3/33=.09). If the red zone is defined as 1-3, the percentage of days that the inventory level is in the red zone is 24%. Therefore the red zone is set to 1-3, since the ratio of 24% to 16% (i.e. 1.5) is lower than the ratio of 16% to 9% (i.e. 1.7). The yellow zone is the range between the green and red zones, that is becomes 4-5.

The resulting ratio between the zones is 12%:64%:24%. For even smaller buffer sizes the allocation of the zones may be performed slightly differently.

In one example, the buffer size is 1. The green zone is an inventory level of 1 and the red zone is an inventory level of 0.

In another example, the buffer size is 2. At least one inventory level is allocated to the green zone. Alternative ways for allocating the zones are: a) Green-green-black for inventory levels 2, 1 and 0 respectively; b) Green-yellow-black for inventory levels 2, 1 and 0 respectively; or c) Green-red-black for inventory levels 2, 1 and 0 respectively. Triggering buffer size recalculation

Each SL-SKU has an upper threshold and a lower threshold. The upper threshold is used to identify when inventory levels are too high and a buffer size decrease should be triggered. The lower threshold is used to identify when inventory levels are too low and a buffer size increase should be triggered.

For each SL-SKU, a buffer size decrease is triggered when the presence of the historical inventory levels within the green zone exceeds a respective upper threshold (also denoted a too much green or TMG condition). A buffer size increase is triggered when the presence of the historical inventory levels within the red zone exceeds a respective lower threshold (also denoted a too much red or TMR condition).

In some embodiments the presence of historical inventory levels within a zone is the penetration of the inventory levels into the zone. As described below, penetration is based on both the number of data points within a zone and the inventory level of each data point. In other embodiments the presence of historical inventory levels within a zone is based only on the number of data points within the zone, without considering the exact level of each data point.

Optionally, a buffer size recalculation is also triggered under one or more of the following conditions: a) When the historical inventory levels remain outside the red zone and the black zone longer than a specified time period a buffer size decrease is triggered. b) When the historical inventory levels increase and/or stay at the same level for longer than a specified time period a buffer size decrease is triggered. c) When the historical inventory levels decrease and/or stay at the same level for longer than a specified time period a buffer size increase is triggered.

Calculating Thresholds for Use in the Significance Test

Optionally, the upper and/or lower thresholds are calculated so that they are suitable for use in a significance test for detecting too much or too little inventory. The significance test is based on a significance level, which is typically input by the end user. The significance levels for establishing thresholds for the TMG and TMR conditions may be the same or different.

An example of calculating a lower threshold to be used in a significance test using estimated inventory data for a particular SL-SKU is now presented. As will be appreciated by the skilled person, a similar analysis may be used to calculate the upper threshold using the same estimated inventory data. Reference is now made to Fig. 6 A, which is an example of estimated inventory levels showing the transitions between black, red and yellow zones. Days in which the inventory levels are in the red and black zones (i.e. below the yellow zone) are indicated by arrows. The legend beneath each arrow indicates the inventory level zone on the following day. R->R and B->R indicate that the inventory level on the following day was in the red zone. The labels R->Y and B->Y both indicate that the inventory level on the following day was in the yellow zone.

For the purpose of this example, assume that the significance level is set at 5% and the probabilities of the inventory zones are 16%:68%:16% (for green, yellow and red/black zones). The probability of the inventory level being in the red (or black) zone in any one day is 16% by definition.

The probability of being in the red (or black) zone when the inventory level of previous day was also in the red (or black) zone is determined from the estimated data. In the present example, the inventory level was in the red or black zones for eight days. For six of these eight days the inventory level on the following day was in the red/black zones (i.e. R->R and B->R). This gives a probability of 6/8, i.e. 75%.

Once these two parameters are known (in this example 16% and 75%), the probability of being in the red (or black) zone in any number of consecutive days may be calculated as shown in Table 3.

Table 3

In embodiments of the invention, when the probability of the inventory level being in the red/black zones is less than the significance level, the level of inventory at that day is not considered “normal”. For example, when the set significance level is 5%, then being in the red/black zones for six consecutive days is not “normal” because on the sixth day the probability drops below 5%.

Alternate embodiments of the invention also take the depth of penetration into the red/black zones into consideration.

As used herein the term “penetration” reflects the position of the inventory level within a zone. A high penetration means that the inventory level is relatively far from the yellow zone range whereas a low penetration means that the inventory level is close to or on the border of the yellow zone. Fluctuations of the inventory levels within the yellow zone may be considered “noise” and do not affect the buffer adjustment.

When the inventory level is on the border line between the yellow and red zone, the penetration is 0%. When the inventory level is zero, the penetration is 100%. Optionally, for a given zone (e.g. red/black or green) the threshold for triggering the buffer update is calculated from the significance level and: a) Probability of entering into the zone (denoted herein Pi); b) Probability of staying in the zone (denoted herein P2); c) Expected zone penetration (denoted herein P3).

Further optionally, the penetration threshold value is calculated as:

Penetration threshold value=P3*(Fog(Significance level/Pi,P2)+l; where Log(a,b) means the logarithm of a in base b.

In the example presented in Table 4, the red zone is defined as inventory levels of one, two and three. Based on the estimated data, the probability of inventory=3 is 51.5%; the probability of inventory=2 is 24.7%; the probability of inventory=l is 18.5%; and the probability of inventory = 0 is 2.0%. The red penetration is 0%, 33%, 66%, and 100% when the inventory levels are 3, 2, 1, and 0, respectively.

For each inventory level, the expected level of penetration is calculated by multiplying the red penetration level with the probability of being at that inventory level. In this example, the penetration at an inventory level of 2 equals 0.082. The total expected penetration into the red zone is the sum of the penetration at all the red/black inventory levels, in this case 0.257 (or 25.7%).

Table 4

Table 5 shows the accumulated red zone penetration over consecutive days. The accumulated penetration increases by 25.7% each day. In this example, a trigger to increase the buffer size would be output when the penetration of the estimated inventory levels into the red/black zone exceeds 130% (because the 5.06% significance is equivalent to 128.5% accumulated penetration).

Table 5

Similar logic may also be applied to trigger buffer size decrease based on green zone penetration.

In the example above, the probability of entering the red zone (Pi) is 16%, the probability of staying in the red zone (P2) is 75% and the expected red zone penetration (P3) is 25.7%. Using the equation given above, for a significance level of 5% the penetration threshold value is: Penetration threshold value=25.7%*(Log(5%/16%,75%)+l=130%;

In another example, the trigger value for accumulated penetration has been calculated as

150% based on the significance level. Given an inventory penetration into the red zone as shown in Table 6:

Table 6

The buffer increase trigger is output at day 5 because 170% >= 150%.

In yet another example, the trigger value for accumulated penetration has been calculated as 150% based on the significance level. Given an inventory penetration into the red zone as shown in Table 7:

Table 7 The buffer increase trigger is output at day 3 as 200% >= 150%.

Once the thresholds are set and buffer management starts, the penetration into the red and green zones is monitored and the accumulated penetration is calculated.

Reference is now made to Fig. 6B, which shows an example of monitoring accumulated penetration into the red zone. Penetration into the red zone begins on day 4 at 10%. Days 5-7 penetrate into the red zone by 20%, 10% and 40% respectively. Thus the accumulated penetration equals 80%. If, for example, the trigger threshold is 35%, a buffer size update will be triggered on day 6.

Buffer increase and decrease percentages Optionally, the buffer size increase and/or decrease parameters are updated based on the ranges of the buffer zones. Further optionally, the increase and/or decrease parameters are calculated from the green, yellow and red zone percentages. The increase and/or decrease parameters may also be based on penetration into the green and red zones at the time of the trigger output.

In a first exemplary embodiment:

Increase percentage=R/2+Y/2; and Decrease percentage=G/2+Y/2, where G equals the green zone percentage, Y equals the yellow zone percentage and R equals the red zone percentage.

In a second exemplary embodiment:

Increase percentage=(Red zone penetration at the time of the trigger detection)+Y/2; and

Decrease percentage=(Green zone penetration at the time of the trigger detection)+Y/2,

Buffer size updating

Optionally, the buffer size is recalculated when a significant change in the inventory level is detected. When a buffer size increase is triggered (TMR condition), the buffer size is increased by the buffer size increase percentage. When a buffer size decrease is triggered (TMG condition), the buffer size is decreased by the buffer size decrease percentage.

Optionally, the buffer size is recalculated by the DBM system and the new buffer size is output to the location. Alternately or additionally, the DBM system outputs the trigger to the location, and the buffer size recalculation is performed at the location. Further optionally, the new buffer size is returned from the location to the DBM system so that the buffer data may be updated.

Replenishment

Optionally, the SL-SKU is replenished when the respective buffer size is increased. Typically, when there are large number of SKUs, for example moving from a warehouse to a store, some SKUs are in red, some are in yellow, some are in green. The operational recommendation is that SL-SKUs in the red zone are handled first, then yellow and then green. When there is an operational limitation (such as the capacity of a truck that carries replenished items, work time, etc.), some SKUs may be handled with higher priority. For example, if an SL- SKU in the green zone is not handled immediately and some of the same SL-SKU are consumed (e.g. sold today at a store), after the system recalculate the priority the buffer status of that SL- SKU may become yellow on the following day and will be handled with higher priority.

User Input -

Optionally, the user inputs information that is used by the DBM system to control the buffer size. This information may be the same for some or all of the SL-SKUs or may differ for each SL-SKU.

User input may include but are not limited to:

1) Significance level - The significance level may be used to determine whether a buffer size increase or decrease should be triggered, as described herein for some embodiments of the invention. Optionally, the significance level is different for the green zone and the red zones.

2) Lead time information - Used to estimate the inventory levels.

3) Spike exclusion - Used to exclude outlier consumption from the estimated inventory level analysis.

4) Seasonality up threshold- Used to exclude seasonal surge in demand from the estimated inventory level analysis.

5) Seasonality down threshold- Used to exclude seasonal drop in demand from the estimated inventory level analysis.

6) Target shortage - Used to determine the range of the black zone.

7) Sigma coverage- Automatically calculated from the user set target shortage and the normal distribution function.

Data Filtering

Optionally, before the inventory levels are estimated the input data is filtered to remove or reduce irregular consumption. Consumption data may consist of two types of consumption, regular consumption and irregular consumption. The irregular consumption may further be divided into spike and seasonal consumption. Optionally, filtering of irregular consumption is based on the values of parameter(s) for data filtering which are input by the client. These parameters may include but are not limited one or more of:

• Spike Exclusion;

• Seasonality up threshold; and

• Seasonality down threshold.

Some embodiments of the invention use feedback and feedforward technology to control the inventory. The feedback part is performed by embodiments of the DBM system described herein. However, there are often cases where the change in demand is very rapid and big as compared to the ability to adjust to the change in demand. One example is demand on a heavy promotion day. For example, a grocery store may discount 50% off the tag price on a particular day for a particular SKU. That information is advertised to attract customers. If the store sets such a special promotion day once a month, then we may know that on the promotion day the SKU sells about five times more than a regular day. In actual operation, the store prepares enough inventory before the promotion, which is meant here by the term “feedforward”. Therefore, in order to determine the parameters for the DBM, such expected consumption on the promotion day may be filtered out before the analysis. The same process may be applied for seasonal products. A characteristic of a seasonal product is a rapid increase in demand and/or a rapid decrease in demand. If we know that increase/decrease pattern from past experience and/or data of same or similar SKUs and if the seasonal change is more rapid than the DBM system is able to follow by feedback, such rapid increase/decrease may be taken into account before starting the analysis used to determine the DBM parameters.

Exemplary Embodiments

Reference is now made to Fig. 7A, which is a simplified flowchart of a method for updating a buffer size for a single SKU at a particular location, according to a first exemplary embodiment of the invention.

In 710-712 the historical data for the SL-SKU is input, the inventory levels are estimated and the zones are calculated within the buffer size, according to embodiments presented herein.

In 713 the trigger thresholds for increasing and decreasing buffer size are calculated. The triggers may be based on the level of penetration of the estimated inventory levels into the red and green zones, according to embodiments presented herein.

In 714 the increase and decrease percentages are calculated. The increase and decrease percentages may be based, at least in part, on the relative distribution of the zones within the buffer, according to embodiments presented herein.

In 715 new inventory data is input (e.g. inventory consumption for the current day) and penetration into the red (or red/black) and green (or green/cyan) zones is calculated, according to embodiments presented herein.

In 716 the accumulated zone penetration is compared to the trigger thresholds calculated in 713, according to embodiments presented herein. If neither of the thresholds is exceeded, the method returns to 715. If one of the thresholds is exceeded, a buffer size update is triggered. In 717 the buffer size is updated according to embodiments presented herein (i.e. increased or decreased based on which threshold was exceeded). The method then returns to 710.

Reference is now made to Fig. 7B, which is a simplified flowchart of a method for updating a buffer size for a single SKU at a particular location, according to a second exemplary embodiment of the invention. Steps 720-727 are substantially similar to 710-717 of Fig. 7A, with the difference that after the buffer size is updated the method returns to 725.

Reference is now made to Fig. 8, which is a simplified schematic diagram of calculating parameters for a DBM system, according to an exemplary embodiment of the invention. The inputs into the DBM systems, outputs from the DBM systems and processes used to calculate all the parameters are labeled.

Process 1 (813) calculates the daily inventory fluctuation as a result of pure replenishment of the consumed amount (814), in accordance with embodiments described herein. The DBM system inputs to process 1 (813) are: a) The historical consumption data (810); b) Delivery frequency (811); and c) Replenishment lead time (812).

Process 1 (813) is similar to the process illustrated in Table 1, while taking into account the delivery frequency as an additional variable affecting the delay between consumption of the SKU and its replenishment.

Process 2 (816) uses the estimated inventory level (814) to calculate the respective red zone, yellow zone and green zone percentages (817), in accordance with embodiments described herein. The red zone, yellow zone and green zone percentages (817) are system outputs and are also used by Process 5 (825) to calculate the increase and decrease percentage parameters (826).

Process 3 (818) inputs the daily inventory fluctuation (814) and red zone, yellow zone and green zone percentages (817) and calculates Pi, P2 and P3 for the red zone and for the green zone (819). Pi, P2 and P3 are respectively the probability of being in a zone, the probability of staying in a zone and the expected zone penetration as described above.

Process 4 (821) uses the respective Pi, P2 and P3 for the red and green zones (819) and the specified significance level (820) to determine the threshold for triggering a buffer size increase (822) and the threshold for triggering a buffer size decrease (824), in accordance with embodiments described herein.

Process 5 (825) uses the red zone, yellow zone and green zone percentages (817) to calculate the increase percentage and decrease percentage (826), in accordance with embodiments described herein. Optionally the increase and/or decrease percentages are provided by the user. Process 5 is not needed when both increase and decrease percentages are provided by the user.

The outputs of the exemplary embodiment are the six DBM system parameters: a) Red zone, yellow zone and green zone percentages (817); b) Increase buffer size threshold (822); c) Decrease buffer size threshold (824); and d) Increase and decrease percentages (826).

Reference is now made to Fig. 9, which is a simplified flowchart of a method for calculating a recommended new buffer size, according to an exemplary embodiment of the invention. The six DBM parameters have already been determined and the significance level has been specified.

In 910, the method begins with the historical inventory level data.

In 920, a significance test is applied to the historical inventory levels. If an analysis of the historical inventory levels do not trigger a buffer size change, in 930 the buffer size is not changed. If an analysis of the historical inventory levels triggers a buffer size change, in 940 a new recommended buffer size is calculated from the current buffer size and the increase/decrease percentage parameters.

Optionally, the buffer is not decreased below a minimum buffer size (950), even when the penetration into the green zone exceeds the significance level.

The descriptions of the various embodiments have been presented for purposes of illustration, but are not intended to be exhaustive or limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terminology used herein was chosen to best explain the principles of the embodiments, the practical application or technical improvement over technologies found in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.

It is expected that during the life of a patent maturing from this application many relevant stock keeping units, locations, ways for estimating inventory levels, ways for calculating penetration into zones and significance tests will be developed and the scope of the term stock keeping unit, locations, SKU, SL-SKU, estimated inventory levels, penetration, significance test and significance level are intended to include all such new technologies a priori.

Whenever a numerical range is indicated herein, it is meant to include any cited numeral (fractional or integral) within the indicated range. The phrases “ranging/ranges between” a first indicate number and a second indicate number and “ranging/ranges from” a first indicate number “to” a second indicate number are used herein interchangeably and are meant to include the first and second indicated numbers and all the fractional and integral numerals therebetween.

It is appreciated that certain features of the invention, which are, for clarity, described in the context of separate embodiments, may also be provided in combination in a single embodiment. Conversely, various features of the invention, which are, for brevity, described in the context of a single embodiment, may also be provided separately or in any suitable subcombination or as suitable in any other described embodiment of the invention. Certain features described in the context of various embodiments are not to be considered essential features of those embodiments, unless the embodiment is inoperative without those elements. Although the invention has been described in conjunction with specific embodiments thereof, it is evident that many alternatives, modifications and variations will be apparent to those skilled in the art. Accordingly, it is intended to embrace all such alternatives, modifications and variations that fall within the spirit and broad scope of the appended claims.

It is the intent of the applicant(s) that all publications, patents and patent applications referred to in this specification are to be incorporated in their entirety by reference into the specification, as if each individual publication, patent or patent application was specifically and individually noted when referenced that it is to be incorporated herein by reference. In addition, citation or identification of any reference in this application shall not be construed as an admission that such reference is available as prior art to the present invention. To the extent that section headings are used, they should not be construed as necessarily limiting. In addition, any priority document(s) of this application is/are hereby incorporated herein by reference in its/their entirety.

Claims

WHAT IS CLAIMED IS:

1. A dynamic buffer management system for inventory control, comprising: a data interface configured for communicating over a network with a plurality of inventory locations, each of said inventory locations stocking a plurality of stock keeping units; and a processing circuitry associated with said data interface, configured to: for each of said stock keeping units at each of said locations: maintain buffer data for said stock keeping unit (SKU), said buffer data comprising a respective buffer size for said SKU; estimate inventory levels based on data input over said data interface; calculate respective ranges of a plurality of zones within said buffer size based on specified ratios for said estimated inventory levels being within each of said zones; and output a trigger for recalculating said buffer size when a presence of historical inventory levels within at least one of said calculated ranges exceeds a respective threshold for said zone and maintain said buffer size unchanged when said presence of said historical inventory levels is within said respective threshold.

2. A dynamic buffer management system according to claim 1, wherein said data for estimating said inventory levels comprises consumption data for said stock keeping unit at said location.

3. A dynamic buffer management system according to claim 1 or claim 2, wherein said processing circuitry is further configured to recalculate said buffer size when said trigger is output and to output said recalculated buffer size to said location over said data interface.

4. A dynamic buffer management system according to any one of claims 1-3, wherein said plurality of zones comprise a first zone indicative of a high level of inventory, a second zone indicative of a target amount of inventory and a third zone indicative of an undersupply of inventory.

5. A dynamic buffer management system according to claim 4, wherein said processing circuitry is configured to output a buffer increase trigger when a time period said historical inventory levels are present in said third zone exceeds a first threshold, said first threshold being calculated from a probability of said estimated inventory levels remaining in said third zone for consecutive time periods.

6. A dynamic buffer management system according to claim 4, wherein said processing circuitry is configured to output a buffer increase trigger when a penetration of said historical inventory levels within said third zone exceeds a second threshold, said first penetration level being calculated from a probability of said estimated inventory levels remaining at each level of said third zone for consecutive time periods.

7. A dynamic buffer management system according to any one of claims 4-6, wherein said processing circuitry is configured to output a buffer decrease trigger when a time period said historical inventory levels are present in said first zone exceeds a third threshold, said third threshold being based on a probability of said estimated inventory levels remaining in said first zone for consecutive time periods.

8. A dynamic buffer management system according to any one of claims 4-6, wherein said processing circuitry is configured to output a buffer decrease trigger when a penetration of said historical inventory levels within said first zone exceeds a fourth threshold, said fourth threshold being based on a probability of said estimated inventory levels remaining at each of level of said first zone for consecutive time periods.

9. A dynamic buffer management system according to any one of claims 1-8, wherein said buffer size is decreased when said historical inventory levels remain outside said third zone longer than a first specified time period.

10. A dynamic buffer management system according to any one of claims 1-9, wherein said buffer size is decreased when said historical inventory levels increase for longer than a second specified time period.

11. A dynamic buffer management system according to any one of claims 1-10, wherein said buffer size is increased when said historical inventory levels decrease for longer than a third specified time period.

12. A dynamic buffer management system according to any one of claims 4-11, wherein a multiple for buffer size increase is based on said recalculated ranges of said second zone and said third zone.

13. A dynamic buffer management system according to any one of claims 4-12, wherein a multiple for buffer size decrease is based on said recalculated ranges of said first zone and said second zone.

14. A dynamic buffer management system according to any one of claims 1-13, wherein said processing circuitry is further configured to identify anomalies in said estimated inventory levels and to output an indicator of said anomalies to said location over said data interface.

15. A method for inventory control, comprising: inputting, from inventory locations stocking a plurality of stock keeping units, respective data for estimating inventory levels of said stock keeping units at said locations over time; for each of said stock keeping units at each of said locations: maintaining buffer data for said stock keeping unit (SKU), said buffer data comprising a respective buffer size for said SKU at said location; estimating inventory levels of said stock keeping unit at said location using said respective input data for said SKU at said location; calculating respective ranges of a plurality of zones within said buffer size based on specified ratios for said estimated inventory levels being within each of said zones; and outputting a trigger for recalculating said buffer size when a presence of historical inventory levels within at least one of said calculated ranges exceeds a specified threshold and maintaining said buffer size unchanged when said presence of said historical inventory levels within said at least one of said recalculated ranges is within said specified threshold.

16. A method for inventory control according to claim 15, wherein said estimated inventory levels are based on consumption data for said stock keeping unit at said location.

17. A method for inventory control according to claim 15 or claim 16, further comprising: recalculating said buffer size when said trigger is output; and outputting said recalculated buffer size to said location.

18. A method for inventory control according to any one of claims 15-17, wherein said plurality of zones comprise a first zone indicative of a high inventory level, a second zone indicative of a target amount of inventory and a third zone indicative of an undersupply of inventory.

19. A method for inventory control according to claim 18, wherein said trigger comprises a buffer increase trigger when a time period said historical inventory levels are present in said third zone exceeds a first threshold, said first threshold being calculated from a probability of said estimated inventory levels remaining in said third zone for consecutive time periods.

20. A method for inventory control according to claim 18, wherein said trigger comprises a buffer increase trigger when a penetration of said historical inventory levels within said third zone exceeds a second threshold, said second threshold being calculated from a probability of said estimated inventory levels remaining at each level of said third zone for consecutive time periods.

21. A method for inventory control according to any one of claims 18-20, wherein said trigger comprises a buffer decrease trigger when a presence of said historical inventory levels within one of said first zone and said second zone exceeds a second time period, said second time period being based on a probability of said estimated inventory levels remaining in one of said first zone and said second zone for consecutive time periods.

22. A method for inventory control according to any one of claims 18-20, wherein said trigger comprises a buffer decrease trigger when a presence of said historical inventory levels within said first zone exceeds a second penetration level, said second penetration level being based on a probability of said estimated inventory levels remaining at each of level of said first zone for consecutive time periods.

23. A method for inventory control according to any one of claims 18-22, further comprising decreasing said buffer size when said historical inventory levels remain outside said third zone longer than a first specified time period.

24. A method for inventory control according to any one of claims 18-23, further comprising decreasing said buffer size when said historical inventory levels increase for longer than a second specified time period.

25. A method for inventory control according to any one of claims 18-24, further comprising increasing said buffer size when said historical inventory levels decrease for longer than a third specified time period.

26. A method for inventory control according to any one of claims 18-25, further comprising calculating a multiple for increasing said buffer size from said recalculated ranges of said second zone and said third zone.

27. A method for inventory control according to any one of claims 18-26, further comprising calculating a multiple for decreasing said buffer size from said recalculated ranges of said first zone and said second zone.

28. A method for inventory control according to any one of claims 15-27, further comprising: identifying anomalies in said estimated inventory levels; and outputting an indicator of said anomalies to said location.

29. A computer program product comprising a program code which, when implemented on a processor, cause the processor to carry out the method of any one of claims 15-28.