Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1-2, 6-14, 16-18 and 20-23 are rejected under 35 U.S.C. 103 as being unpatentable over Spector et al. (US2018/0315319 A1 – hereinafter Spector) and further in view of Prokhorenkova et al. (“CatBoost: Unbiased Boosting with Categorical Features” – hereinafter Prokhorenkova).
In regards to claim 1, Spector discloses a method comprising:
receiving logistics operation order data, wherein the logistics operation order data identifies at least one logistics operation to be performed (Spector fig. 2a element 200 receives an order and para. [0055] cites “The received orders 182 may be assigned an order identifier, stored in the order storage of the routing system 120 and may include the data associated with the order, including, for example, the origination location and destination location and any order criteria or data such as any pick-up time window or delivery time window, an order status (e.g., new, assigned, picked up or not, in-transit, delivered or completed, etc.)…”. There is logistic data associated each order and operation to be performed, delivery.) with a plurality of logistics characteristics including at least a shipment type, shipment time priority and a shipment location (Spector fig. 2a element 200 receives an order and para. [0055] cites “The received orders 182 may be assigned an order identifier, stored in the order storage of the routing system 120 and may include the data associated with the order, including, for example, the origination location and destination location and any order criteria or data such as any pick-up time window or delivery time window, an order status (e.g., new, assigned, picked up or not, in-transit, delivered or completed, etc.), an identifier for any driver assigned to the order, or other information, a distance between the pick-up location and the destination location, including the weight of the item to be delivered, any time windows or times associated with the pick-up or delivery of the order, any necessary or desired equipment to be used for the order (for example, access cards, lift gates on a vehicle, etc.), and necessary or desired compliance or training for the driver (e.g., Health Insurance Portability and Accountability Act (HIPAA) compliance or the like), any vehicle size needs (e.g., van or truck), any requirements for completion of any order (e.g., signature at delivery or picture of delivery) and any other remarks which may needed or helpful for completion of the order (for example, gate codes, identification of entrances at the destination location, a phone number to call at delivery, etc.).”. The examiner interprets shipment type to vehicle size needs and any requirements such signature at delivery or picture of delivery, shipment time priority is the time windows for delivery and shipment location is destination location.)
training the one or more machine learning algorithms with historical logistics data; (Spector para. [0014] teaches using machine learning by delivery system and para. [0017] teaches training the machine learning model based historical order data and para. [0055] teach order has logistic data associated with it.)
executing the one or more trained machine learning algorithms to predict a logistics provider to perform the at least one logistics operation at a given level of performance (Spector fig. 2B element 252, 254, 256, 259, and 260 disclose selecting a driver (logistics provider) to perform the one logistic operation (delivering order), wherein a user is selected, notified of order deliver wherein they can accept or rejected it, and if accepted the order is assigned to the driver for delivery. Also, Spector para. [0015] teaches removing drivers would have low possibility of being a good and delivering the package and para. [0020] teaches selecting the highest rank driver who has the best likelihood of delivery, wherein given performance level is delivering the package.)
wherein the steps of the method are executed by a processing device operatively coupled to a memory. (Spector fig. 1 discloses a system that performs the operations above and had memory in driver data storage, model storage, rule storage, order storage and so on.)
However Spector does not explicitly disclose wherein the executing of the one or more trained machine learning algorithms further comprises executing a first trained machine learning algorithm, wherein the first trained machine learning algorithm includes a plurality of sequential classifiers that respectively correspond to the plurality of logistics characteristics of the at least one logistics operation, wherein each sequential classifier is trained using data from a previously trained sequential classifier and corrects any identified errors from the previously trained sequential classifier, and wherein the first trained machine learning algorithm aggregates predictions made by each of the sequential classifiers to determine the logistics provider.
Prokhorenkova discloses wherein the executing of the one or more trained machine learning algorithms further comprises executing a first trained machine learning algorithm, wherein the first trained machine learning algorithm includes a plurality of sequential classifiers, wherein each sequential classifier is trained using data from a previously trained sequential classifier and corrects any identified errors from the previously trained sequential classifier, and wherein the first trained machine learning algorithm aggregates predictions made by each of the sequential classifiers. (Prokhorenkova section 3.1 last paragraph teaches using CatBoost and section 4.2 teaches sequential decision trees in CatBoost and algorithm 2 and page 6-7 section “Building a tree” teaches a first trained machine algorithm (CatBoost) includes a plurality of sequential classifiers (decision trees), wherein each sequential classifier is trained using data from a previously trained sequential classifier and corrects any identified errors from the previously trained sequential classifier, and wherein the first trained machine learning algorithm aggregates predictions made by each of the sequential classifiers. (See page 6-7 section “building a tree” and Choosing leaf values” wherein it teaches gradients based on previous examples which is residuals (error) of previous tree, thus correcting errors of prior trees and page 16 algorithm 3 line 15 gives an equation for summing all trees to get an output or prediction.)
It would have been obvious to one of ordinary skill in the art before the earliest effective filing date of the claimed invention to modify the teachings of the Spector that deal with logistics operations and uses decision trees with that the teachings of Prokhorenkova’s CatBoost in order to allow for CatBoost for using sequential weak learners (decision trees) to combine and provide an prediction for logistics as both references use decision trees and the motivation to do so is based on Spector disclose in para. [0095] which states that machine learning model may be a decision tree and Prokhorenkova discloses on page 1 last paragraph and on page 2 fourth paragraph that CatBoost is a form of gradient boosting using binary decision trees that outperforms other gradient boosted decision trees, thus the user would be motivated to do so in order to create the best performing system possible.
In regards to claim 2, Spector in view of Prokhorenkova discloses the method of claim 1 further comprising generating, based at least in part on the predicting, a request for the logistics provider to perform the at least one logistics operation, wherein the request is transmitted to the logistics provider. (Spector fig. 2B element 252, 254, 256, 259, and 260 disclose selecting a driver (logistics provider) to perform the one logistic operation (delivering order), wherein a user is selected, notified of order deliver wherein they can accept or rejected it, and if accepted the order is assigned to the driver for delivery.)
In regards to claim 6, Spector in view of Prokhorenkova discloses the method of claim 3 wherein: the historical logistics data specifies one or more features associated with respective ones of a plurality of logistics operations; and (Spector para. [0055] cites “The received orders 182 may be assigned an order identifier, stored in the order storage of the routing system 120 and may include the data associated with the order, including, for example, the origination location and destination location and any order criteria or data such as any pick-up time window or delivery time window, an order status (e.g., new, assigned, picked up or not, in-transit, delivered or completed, etc.)…”. Spector para. [0014] teaches using machine learning by delivery system and para. [0017] teaches training the machine learning model based historical order data. In para. [0055] order have feature data associated logistic operations, destination location, origin location , order status and so on. These are categorical data or feature data associated logistic operation.) the one or more machine learning algorithms automatically encode the one or more features for use in a training dataset. (Prokhorenkova section 3.1 teaches using CatBoost that includes encoding features of training data using one-hot coding.)
In regards to claim 7, Spector in view of Prokhorenkova discloses the method of claim 3 wherein: the historical logistics data specifies one or more features associated with respective ones of a plurality of logistics operations; and (Spector para. [0055] cites “The received orders 182 may be assigned an order identifier, stored in the order storage of the routing system 120 and may include the data associated with the order, including, for example, the origination location and destination location and any order criteria or data such as any pick-up time window or delivery time window, an order status (e.g., new, assigned, picked up or not, in-transit, delivered or completed, etc.)…”. Spector para. [0014] teaches using machine learning by delivery system and para. [0017] teaches training the machine learning model based historical order data. In para. [0055] order have feature data associated logistic operations, destination location, origin location , order status and so on. These are categorical data or feature data associated logistic operation.) and the method further comprises extracting one or more sub-features from the one or more features to be used during the training. (Prokhorenkova page 2-3 section 3.1 teaches category is feature which is represent by a node of a decision tree and it splits into sub features as it progresses down the decision tree and the category is split in sub features.)
In regards to claim 8, Spector discloses the method of claim 3 wherein the one or more machine learning algorithms comprise a decision tree (Spector para. [0095]) trained on historical data (Spector para. [0014] teaches using machine learning by delivery system and para. [0017] teaches training the machine learning model based historical order data and para. [0055] teach order has logistic data associated with it.) and the plurality of decision trees are respectively trained with different portions of the historical logistics data. (Prokhorenkova fig. 1 shows an algorithm 1 shows where in a first model (decision tree) is trained using examples 1-5, and second model is trained with examples 1-6 and third with be examples 1-7. Also see section 4.2 on page 5 that explains this further. )
In regards to claim 9, Spector in view of Prokhorenkova discloses the method of claim 8 wherein the analyzing comprises:
sequentially analyzing the logistics operation order data with respective ones of the plurality of decision trees to generate respective predictions; and (Prokhorenkova section 4.2 teaches sequential decision trees and Spector teaches using logistic operation order data.)
aggregating the respective predictions to determine the logistics provider to perform the at least one logistics operation.(Prokhorenkova page 1 section 1 first paragraph teaches combining weak learner to create a strong prediction, this would be aggregating the respected predictions for the weak learners.)
In regards to claim 10, Spector in view of Prokhorenkova discloses the method of claim 3 further comprising harvesting the historical logistics data from at least one of a customer relationship management system, an enterprise resource planning system, a sales system and an order fulfillment system. (Spector para. [0024] teaches “In one embodiment, filtering the set of drivers to determine a reduced set of drivers comprises applying a set of filtering rules before applying the filtering model and wherein the filtering model is a recurrent neural network model trained on a set of historical snapshot of previously completed orders.” As the historical data is based on previously completed order the examiner interprets this to be an order fulfillment system.)
In regards to claim 11, Spector in view of Prokhorenkova discloses the method of claim 1 wherein the one or more machine learning algorithms comprise ensemble decision tree-based boosting algorithm. (Spector para. [0095] disclose the machine learning model is a decision tree and Prokhorenkova abstract and page 1 section 1 teaches using CatBoost which is a form gradient categorical boosting using decision trees that out performs existing state of the art gradient boosted decision trees. It also teaches gradient boosting is an ensemble of predictions, wherein the predictions are weak base predictions or decision trees. Thus, it teaches using an ensemble decision tree boosting algorithm.)
In regards to claim 12, Spector in view of Prokhorenkova disclose the method of claim 11 wherein the one or more machine learning algorithms comprise a categorical boosting algorithm. (Prokhorenkova page 1 last paragraph teaches using CatBoost which categorial boosting, thus it a categorical boosting algorithm.)
In regards to claim 13, Spector in view of Prokhorenkova discloses the method of claim 1 wherein the one or more machine learning algorithms comprise a shallow learning algorithm. (Spector para. [0095] teaches the machine learning model by a decision tree which a type of shallow learning algorithm.)
In regards to claim 14, it is the apparatus embodiment of claim 1 with similar limitations and thus rejected using the same reasoning found in claim 1. It includes the limitations of a process and memory being used which is disclosed in Spector fig. 1 and para. [0117].
In regards to claim 16, it is the apparatus embodiment of claim 8 with similar limitations and thus rejected using the same reasoning found in claim 8.
In regards to claim 17, it is the apparatus embodiment of claim 9 with similar limitations and thus rejected using the same reasoning found in claim 9.
In regards to claim 18, it is the non-transitory process-readable storage medium embodiment of claim 1 with similar limitations and thus rejected using the same reasoning found in claim 1.
In regards to claim 20, it is the non-transitory process-readable storage medium embodiment of claim 8 with similar limitations and thus rejected using the same reasoning found in claim 8.
In regards to claim 21, it is the non-transitory process-readable storage medium embodiment of claim 11 with similar limitations and thus rejected using the same reasoning found in claim 11.
In regards to claim 22, it is the non-transitory process-readable storage medium embodiment of claim 12 with similar limitations and thus rejected using the same reasoning found in claim 12.
In regards to claim 23, it is the non-transitory process-readable storage medium embodiment of claim 13 with similar limitations and thus rejected using the same reasoning found in claim 13.
Claims 4-5 are rejected under 35 U.S.C. 103 as being unpatentable over Spector et al. (US2018/0315319 A1 – hereinafter Spector) in view of Prokhorenkova et al. (“CatBoost: Unbiased Boosting with Categorical Features” – hereinafter Prokhorenkova). as applied to claim 3 above, and further in view of Kuettner (US 2020/0160263 A1).
In regards to claim 4, Spener in view of Prokhorenkova discloses the method of claim 3 but does disclose wherein the historical logistics data specifies a plurality of logistics operations associated with respective ones of a plurality of logistics providers, and whether there were any issues with respective ones of the plurality of logistics operations.
Keuttner discloses historical logistic data associated with a logistic provider and whether there were any issues with respective one of logistic operations. (Keuttner para. [0095] discloses wherein a record of each deliver driver is maintain and show issues of how many food order that not validate, took too long, or if the driver received complaints.)
It would have been obvious to one of ordinary skill in the art before effective filing date of the claimed invention to modify the teachings of Spencer with that of Keuttner as both references deal with delivery things and it provides the benefit of keeping a record of driver performance to avoid using drivers with a history of bad performance which could hurt the company’s reputation as suggested in paragraph [0095] of Keuttner.
In regards to claim 5, Spencer in view of Prokhorenkova in view of Keuttner disclose the method of claim 4 wherein the historical logistics data specifies one or more features associated with the respective ones of the plurality of logistics operations, and wherein the one or more features include at least one of date, customer, product, product part, logistics operation type, location and cost level. (Spector fig. 2a element 200 receives an order and para. [0055] cites “The received orders 182 may be assigned an order identifier, stored in the order storage of the routing system 120 and may include the data associated with the order, including, for example, the origination location and destination location and any order criteria or data such as any pick-up time window or delivery time window, an order status (e.g., new, assigned, picked up or not, in-transit, delivered or completed, etc.)…”.)
Response to Arguments
Applicant's arguments filed 11 May 2026 have been fully considered but are persuasive in regards to the rejection under 35 USC 101 for being abstract idea and 35 USC 102 are persuasive. As such the rejection under 35 USC 101 and 102 are withdrawn. However, upon further consideration, a new ground(s) of rejection is made in view of Spector et al. (US2018/0315319 A1 – hereinafter Spector) and further in view of Prokhorenkova et al. (“CatBoost: Unbiased Boosting with Categorical Features” – hereinafter Prokhorenkova).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to PAULINHO E SMITH whose telephone number is (571)270-1358. The examiner can normally be reached Mon-Fri. 10AM-6PM CST.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Abdullah Kawsar can be reached at 571-270-3169. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/PAULINHO E SMITH/Primary Examiner, Art Unit 2127
Prosecution Insights