The document describes an algorithm developed for a client called Global Links to estimate the number of shipping containers needed for deliveries. The algorithm uses stochastic rotation and machine learning techniques to iteratively improve its packing of inventory items into containers. It outputs statistics on the estimated number of containers, their dimensions, and error metrics to help the client plan shipments more accurately.
The document describes how to interpret an Agile burn-down chart as a tool for earned value management. It provides examples of setting up baseline burn-down charts for a project with 3 backlog items totaling 120 hours. A series of exercises demonstrates updating the burn-down charts as items are completed, including adding a new 4th item. The appendix summarizes that the project completed 4 items totaling 160 planned hours using 140 actual hours, resulting in an efficiency of 114%.
Email marketing for charities and non-profitsMailNinja
The document discusses several ways that charities and non-profits can improve their email marketing efforts. It suggests that they often fail to clearly identify the benefits of signing up, properly welcome new subscribers, tailor content to individual subscribers, and follow up on incomplete donations. The document provides recommendations for charities to start simple by defining email's role and evaluating past performance, and to use email data to inform audience segmentation and targeting in order to boost fundraising through more effective email campaigns. MailNinja offers services to help charities fine-tune their email messaging, design, and analytics.
S. Mark Maitland is seeking a position as a business manager, controller, or accountant in the private or public sector. He has over 30 years of experience in accounting, financial management, and personnel administration. He holds a B.S. in Business Administration with a focus on management and accounting. Maitland has worked in accounting and financial management roles for various companies, government agencies, and non-profits.
This document provides an introduction to commonly used features in PowerPoint. It will demonstrate how to add and format slides, insert graphics, videos and other media, apply transitions between slides, and add interactive elements like buttons. The goal is to serve as both a user guide and practice material for learning PowerPoint.
Doug Arent has over 20 years of experience in leadership, operations management, training, and logistics. He holds an MBA and MA from the University of Redlands, and a BS from Southern Illinois University. Currently he is an adjunct faculty member at the University of Redlands, where he has taught leadership and organizational behavior courses since 2011. Prior experience includes positions in training, operations management, logistics, and maintenance management for the U.S. Marine Corps, Tapestry Solutions, Terex Utilities, and Abbott Vascular.
WVC Living Laboratory and Outdoor ClassroomCarol Kuster
The West Valley College Vasona Creek Project aims to create a living laboratory and outdoor classroom along Vasona Creek to enhance education across disciplines and inspire innovative solutions to complex environmental problems. The plan is for the creek area to serve as a natural resource for the college to lay the groundwork for creative solutions that offer hope for the future of natural areas and inspire students through hands-on learning outdoors.
The document describes how to interpret an Agile burn-down chart as a tool for earned value management. It provides examples of setting up baseline burn-down charts for a project with 3 backlog items totaling 120 hours. A series of exercises demonstrates updating the burn-down charts as items are completed, including adding a new 4th item. The appendix summarizes that the project completed 4 items totaling 160 planned hours using 140 actual hours, resulting in an efficiency of 114%.
Email marketing for charities and non-profitsMailNinja
The document discusses several ways that charities and non-profits can improve their email marketing efforts. It suggests that they often fail to clearly identify the benefits of signing up, properly welcome new subscribers, tailor content to individual subscribers, and follow up on incomplete donations. The document provides recommendations for charities to start simple by defining email's role and evaluating past performance, and to use email data to inform audience segmentation and targeting in order to boost fundraising through more effective email campaigns. MailNinja offers services to help charities fine-tune their email messaging, design, and analytics.
S. Mark Maitland is seeking a position as a business manager, controller, or accountant in the private or public sector. He has over 30 years of experience in accounting, financial management, and personnel administration. He holds a B.S. in Business Administration with a focus on management and accounting. Maitland has worked in accounting and financial management roles for various companies, government agencies, and non-profits.
This document provides an introduction to commonly used features in PowerPoint. It will demonstrate how to add and format slides, insert graphics, videos and other media, apply transitions between slides, and add interactive elements like buttons. The goal is to serve as both a user guide and practice material for learning PowerPoint.
Doug Arent has over 20 years of experience in leadership, operations management, training, and logistics. He holds an MBA and MA from the University of Redlands, and a BS from Southern Illinois University. Currently he is an adjunct faculty member at the University of Redlands, where he has taught leadership and organizational behavior courses since 2011. Prior experience includes positions in training, operations management, logistics, and maintenance management for the U.S. Marine Corps, Tapestry Solutions, Terex Utilities, and Abbott Vascular.
WVC Living Laboratory and Outdoor ClassroomCarol Kuster
The West Valley College Vasona Creek Project aims to create a living laboratory and outdoor classroom along Vasona Creek to enhance education across disciplines and inspire innovative solutions to complex environmental problems. The plan is for the creek area to serve as a natural resource for the college to lay the groundwork for creative solutions that offer hope for the future of natural areas and inspire students through hands-on learning outdoors.
This document provides an overview of logistic regression, including:
- Logistic regression models the log-odds of an outcome as a linear combination of predictors.
- Maximum likelihood estimation is commonly used to estimate the coefficients by maximizing the log-likelihood.
- There are many implementation decisions regarding goals, data handling, and desired outputs that impact the appropriate modeling approach. Regularization is also an important consideration.
Comparison of Dynamic Programming Algorithm and Greedy Algorithm on Integer K...faisalpiliang1
At this time the delivery of goods to be familiar because the use of delivery of goods services greatly facilitate customers. PT Post Indonesia is one of the delivery of goods. On the delivery of goods, we often encounter the selection of goods which entered first into the transportation and held from the delivery. At the time of the selection, there are Knapsack problems that require optimal selection of solutions. Knapsack is a place used as a means of storing or inserting an object. The purpose of this research is to know how to get optimal solution result in solving Integer Knapsack problem on freight transportation by using Dynamic Programming Algorithm and Greedy Algorithm at PT Post Indonesia Semarang. This also knowing the results of the implementation of Greedy Algorithm with Dynamic Programming Algorithm on Integer Knapsack problems on the selection of goods transport in PT Post Indonesia Semarang by applying on the mobile application. The results of this research are made from the results obtained by the Dynamic Programming Algorithm with total weight 5022 kg in 7 days. While the calculation result obtained by Greedy Algorithm, that is total weight of delivery equal to 4496 kg in 7 days. It can be concluded that the calculation results obtained by Dynamic Programming Algorithm in 7 days has a total weight of 526 kg is greater when compared with Greedy Algorithm.
Recent developments on SMT solvers for non-linear polynomial constraints have become crucial to make the template-based (or constraint-based) method for program analysis effective in practice. Moreover, using Max-SMT (its optimization version) is the key to extend this approach to develop an automated compositional program verification method based on generating conditional inductive invariants. We build a bottom-up program verification framework that propagates preconditions of small program parts as postconditions for preceding program parts and can recover from failures when some precondition is not proved. These techniques have successfully been implemented within the VeryMax tool which currently can check safety, reachability and termination properties of C++ code. In this talk we will provide an overview of the Max-SMT solving techniques and its application to compositional program analysis.
Applying Linear Optimization Using GLPKJeremy Chen
A brief introduction to linear optimization with a focus on applying it with the high-quality open-source solver GLPK.
Originally prepared for an intra-department sharing session.
Making your code faster cython and parallel processing in the jupyter notebookPyData
This document discusses using Cython and parallel processing in Jupyter notebooks to make code faster. It describes using Euler's method to approximate the function y=x^2 for a million points and determine the minimum step size such that the result is within 1e-5 of the correct answer. The author aims to show how to optimize this problem using Cython and parallel processing without financial interests or being a computer scientist.
A.I. for Dynamic Decisioning under Uncertainty (for real-world problems in Re...Ashwin Rao
The document discusses artificial intelligence techniques for dynamic decision making under uncertainty and provides examples from retail and financial trading. It introduces the framework of stochastic control, which involves optimization over time with uncertain and evolving variables. The key problems discussed are inventory control in retail and portfolio optimization in finance. Both are characterized as Markov decision processes with states, actions, rewards, and transitions between states governed by probabilities. The document outlines solutions to basic single-period versions and then more complex multi-period versions of these problems, highlighting the challenges of large and complex state and action spaces as well as delayed or unknown consequences of actions.
IRJET- Solving Quadratic Equations using C++ Application ProgramIRJET Journal
1) The document describes a C++ application program developed to solve quadratic equations. The program uses methods like factoring, completing the square, and the quadratic formula to find the solutions.
2) Field testing of the program showed students using it had an average score of 82.8% on a quadratic equations assessment, demonstrating the program's effectiveness.
3) Advantages of using such an application include reducing errors, supporting problem-solving processes, and creating awareness of mathematical concepts. It allows students to easily test conjectures and replay problem-solving steps.
Recommendation Systems in banking and Financial ServicesAndrea Gigli
The document discusses recommendation systems in banking and financial services. It describes different types of recommendation systems including content-based filtering, collaborative filtering, and hybrid filtering. It then discusses how recommendation systems could be useful in banking by using a bipartite graph and word embedding approaches to represent customer and asset data and identify relationships between them. Code examples are provided for implementing some of these recommendation system techniques.
This document discusses quantum computing business in the Japanese market. It introduces MDR Inc., a Japanese quantum computing startup founded in 2008. MDR develops full-stack quantum computing, from software to hardware, and works with over 20 clients in industries like banking, automotive, materials and more. Some applications discussed include quantum simulation, optimization, and machine learning. The document also provides an overview of the quantum computing developer community and ecosystem in Japan.
The document discusses how lean startup methods can save time and money compared to traditional approaches. It explains that lean startup involves making assumptions about a business plan and then running cheap experiments to validate the assumptions. This allows companies to learn quickly which assumptions are incorrect and adapt the plan accordingly. The document provides an example comparing the expected costs of a project with and without using lean startup experiments to validate assumptions. It finds that running only the most useful experiments can significantly reduce the total expected costs from wasted effort on invalidated assumptions.
Banque de France's Workshop on Granularity: Basile Grassi's slides, June 2016 Soledad Zignago
The document analyzes the impact of large firm dynamics on business cycle fluctuations. It develops a quantitative model where aggregate fluctuations arise solely from firm-level shocks. The model features a finite number of heterogeneous firms that differ in productivity levels. It derives the law of motion for the firm size distribution and shows that aggregate output generated from the model is endogenously persistent, volatile and exhibits time-varying variance. Theoretical results characterize the evolution of the aggregate state and firm productivity distribution dynamics. Quantitatively, the model finds that large firm dynamics account for about one-fourth of aggregate fluctuations.
This document discusses basic probability concepts including probability, events, sample spaces, and counting rules. It defines probability as the chance an uncertain event will occur between 0 and 1. Simple probability is the probability of a single event, while joint probability is the probability of two or more events occurring together. Conditional probability is the probability of one event given another has occurred. The document provides examples of calculating probabilities using formulas and contingency tables. It also covers independence, addition rules, and counting rules for determining the number of possible outcomes.
Logistic Regression in Python | Logistic Regression Example | Machine Learnin...Edureka!
** Python Data Science Training : https://www.edureka.co/python **
This Edureka Video on Logistic Regression in Python will give you basic understanding of Logistic Regression Machine Learning Algorithm with examples. In this video, you will also get to see demo on Logistic Regression using Python. Below are the topics covered in this tutorial:
1. What is Regression?
2. What is Logistic Regression?
3. Why use Logistic Regression?
4. Linear vs Logistic Regression
5. Logistic Regression Use Cases
6. Logistic Regression Example Demo in Python
Subscribe to our channel to get video updates. Hit the subscribe button above.
Machine Learning Tutorial Playlist: https://goo.gl/UxjTxm
This lecture covers the basics of a boolean expression, truth table, and logic diagram. And how they are interconvertible. DLD terms: Minterms, Maxterms, Sum of product(SOP), Canonical sum of products (CSOP), Minimal sum of products(MSOP), and product of sum(POS) are discussed.
WISS 2015 - Machine Learning lecture by Ludovic Samper Antidot
Machine Learning Tutorial
- Study a classical task in Machine Learning : text classification - - Show scikit-learn.org Python machine learning library
- Follow the “Working with text data” tutorial :
http://scikit-learn.org/stable/tutorial/text_analytics/ working_with_text_data.html
- Additional material on http://blog.antidot.net/
SiriusCon 2017 - Sirius-powered Risk Modelling and Simulation in Industry 4.0...Obeo
Our talk reports about how efficient Sirius was in supporting various tasks like creating the graphical editor, adding specific views and interfacing with the underlying simulation engine as well as a report generator. Our tooling was developed in the scope of a 2-year joined project with the Fraunhofer IPT and the IFU cybernetic lab of RWTH Aachen. It also involved Belgian and German end-user companies. It is now being transferred to more companies in the logistics and mechanical engineering domains. It is also available under a permissive Open Source license for all interested companies.
Continuation of Interpolative Scope and Instancing Dialogue... Development of Tools for Instances. and Testing, we are looking in v1 for the Load balancing equation... which we will state in v2 where we will make functional Keys
This document summarizes recent developments in scikit-learn, an open-source machine learning library for Python. It discusses improvements made in version 0.18, including new cross-validation objects and using randomized PCA instead of standard PCA. Upcoming improvements mentioned include adding memory caching to pipelines, a new SAGA solver for logistic regression, and quantile and local outlier factor transformers. It also discusses the scikit-learn user base of 350,000 returning users, its role as core Python infrastructure, and funding and contributions from various academic institutions that support its continued development.
SAT based planning for multiagent systemsRavi Kuril
Multi-agent Classical planning using SAT approach. This document describes the approach and discusses all the experiments and the respective results. I have considered State of the art tools for comparison purpose. Implementation code can be found on GitHub link https://github.com/ravikuril/SATbasedClassicalPlanning . For more Information contact me on ravikuril.du.or@gmail.com
This document provides an overview of logistic regression, including:
- Logistic regression models the log-odds of an outcome as a linear combination of predictors.
- Maximum likelihood estimation is commonly used to estimate the coefficients by maximizing the log-likelihood.
- There are many implementation decisions regarding goals, data handling, and desired outputs that impact the appropriate modeling approach. Regularization is also an important consideration.
Comparison of Dynamic Programming Algorithm and Greedy Algorithm on Integer K...faisalpiliang1
At this time the delivery of goods to be familiar because the use of delivery of goods services greatly facilitate customers. PT Post Indonesia is one of the delivery of goods. On the delivery of goods, we often encounter the selection of goods which entered first into the transportation and held from the delivery. At the time of the selection, there are Knapsack problems that require optimal selection of solutions. Knapsack is a place used as a means of storing or inserting an object. The purpose of this research is to know how to get optimal solution result in solving Integer Knapsack problem on freight transportation by using Dynamic Programming Algorithm and Greedy Algorithm at PT Post Indonesia Semarang. This also knowing the results of the implementation of Greedy Algorithm with Dynamic Programming Algorithm on Integer Knapsack problems on the selection of goods transport in PT Post Indonesia Semarang by applying on the mobile application. The results of this research are made from the results obtained by the Dynamic Programming Algorithm with total weight 5022 kg in 7 days. While the calculation result obtained by Greedy Algorithm, that is total weight of delivery equal to 4496 kg in 7 days. It can be concluded that the calculation results obtained by Dynamic Programming Algorithm in 7 days has a total weight of 526 kg is greater when compared with Greedy Algorithm.
Recent developments on SMT solvers for non-linear polynomial constraints have become crucial to make the template-based (or constraint-based) method for program analysis effective in practice. Moreover, using Max-SMT (its optimization version) is the key to extend this approach to develop an automated compositional program verification method based on generating conditional inductive invariants. We build a bottom-up program verification framework that propagates preconditions of small program parts as postconditions for preceding program parts and can recover from failures when some precondition is not proved. These techniques have successfully been implemented within the VeryMax tool which currently can check safety, reachability and termination properties of C++ code. In this talk we will provide an overview of the Max-SMT solving techniques and its application to compositional program analysis.
Applying Linear Optimization Using GLPKJeremy Chen
A brief introduction to linear optimization with a focus on applying it with the high-quality open-source solver GLPK.
Originally prepared for an intra-department sharing session.
Making your code faster cython and parallel processing in the jupyter notebookPyData
This document discusses using Cython and parallel processing in Jupyter notebooks to make code faster. It describes using Euler's method to approximate the function y=x^2 for a million points and determine the minimum step size such that the result is within 1e-5 of the correct answer. The author aims to show how to optimize this problem using Cython and parallel processing without financial interests or being a computer scientist.
A.I. for Dynamic Decisioning under Uncertainty (for real-world problems in Re...Ashwin Rao
The document discusses artificial intelligence techniques for dynamic decision making under uncertainty and provides examples from retail and financial trading. It introduces the framework of stochastic control, which involves optimization over time with uncertain and evolving variables. The key problems discussed are inventory control in retail and portfolio optimization in finance. Both are characterized as Markov decision processes with states, actions, rewards, and transitions between states governed by probabilities. The document outlines solutions to basic single-period versions and then more complex multi-period versions of these problems, highlighting the challenges of large and complex state and action spaces as well as delayed or unknown consequences of actions.
IRJET- Solving Quadratic Equations using C++ Application ProgramIRJET Journal
1) The document describes a C++ application program developed to solve quadratic equations. The program uses methods like factoring, completing the square, and the quadratic formula to find the solutions.
2) Field testing of the program showed students using it had an average score of 82.8% on a quadratic equations assessment, demonstrating the program's effectiveness.
3) Advantages of using such an application include reducing errors, supporting problem-solving processes, and creating awareness of mathematical concepts. It allows students to easily test conjectures and replay problem-solving steps.
Recommendation Systems in banking and Financial ServicesAndrea Gigli
The document discusses recommendation systems in banking and financial services. It describes different types of recommendation systems including content-based filtering, collaborative filtering, and hybrid filtering. It then discusses how recommendation systems could be useful in banking by using a bipartite graph and word embedding approaches to represent customer and asset data and identify relationships between them. Code examples are provided for implementing some of these recommendation system techniques.
This document discusses quantum computing business in the Japanese market. It introduces MDR Inc., a Japanese quantum computing startup founded in 2008. MDR develops full-stack quantum computing, from software to hardware, and works with over 20 clients in industries like banking, automotive, materials and more. Some applications discussed include quantum simulation, optimization, and machine learning. The document also provides an overview of the quantum computing developer community and ecosystem in Japan.
The document discusses how lean startup methods can save time and money compared to traditional approaches. It explains that lean startup involves making assumptions about a business plan and then running cheap experiments to validate the assumptions. This allows companies to learn quickly which assumptions are incorrect and adapt the plan accordingly. The document provides an example comparing the expected costs of a project with and without using lean startup experiments to validate assumptions. It finds that running only the most useful experiments can significantly reduce the total expected costs from wasted effort on invalidated assumptions.
Banque de France's Workshop on Granularity: Basile Grassi's slides, June 2016 Soledad Zignago
The document analyzes the impact of large firm dynamics on business cycle fluctuations. It develops a quantitative model where aggregate fluctuations arise solely from firm-level shocks. The model features a finite number of heterogeneous firms that differ in productivity levels. It derives the law of motion for the firm size distribution and shows that aggregate output generated from the model is endogenously persistent, volatile and exhibits time-varying variance. Theoretical results characterize the evolution of the aggregate state and firm productivity distribution dynamics. Quantitatively, the model finds that large firm dynamics account for about one-fourth of aggregate fluctuations.
This document discusses basic probability concepts including probability, events, sample spaces, and counting rules. It defines probability as the chance an uncertain event will occur between 0 and 1. Simple probability is the probability of a single event, while joint probability is the probability of two or more events occurring together. Conditional probability is the probability of one event given another has occurred. The document provides examples of calculating probabilities using formulas and contingency tables. It also covers independence, addition rules, and counting rules for determining the number of possible outcomes.
Logistic Regression in Python | Logistic Regression Example | Machine Learnin...Edureka!
** Python Data Science Training : https://www.edureka.co/python **
This Edureka Video on Logistic Regression in Python will give you basic understanding of Logistic Regression Machine Learning Algorithm with examples. In this video, you will also get to see demo on Logistic Regression using Python. Below are the topics covered in this tutorial:
1. What is Regression?
2. What is Logistic Regression?
3. Why use Logistic Regression?
4. Linear vs Logistic Regression
5. Logistic Regression Use Cases
6. Logistic Regression Example Demo in Python
Subscribe to our channel to get video updates. Hit the subscribe button above.
Machine Learning Tutorial Playlist: https://goo.gl/UxjTxm
This lecture covers the basics of a boolean expression, truth table, and logic diagram. And how they are interconvertible. DLD terms: Minterms, Maxterms, Sum of product(SOP), Canonical sum of products (CSOP), Minimal sum of products(MSOP), and product of sum(POS) are discussed.
WISS 2015 - Machine Learning lecture by Ludovic Samper Antidot
Machine Learning Tutorial
- Study a classical task in Machine Learning : text classification - - Show scikit-learn.org Python machine learning library
- Follow the “Working with text data” tutorial :
http://scikit-learn.org/stable/tutorial/text_analytics/ working_with_text_data.html
- Additional material on http://blog.antidot.net/
SiriusCon 2017 - Sirius-powered Risk Modelling and Simulation in Industry 4.0...Obeo
Our talk reports about how efficient Sirius was in supporting various tasks like creating the graphical editor, adding specific views and interfacing with the underlying simulation engine as well as a report generator. Our tooling was developed in the scope of a 2-year joined project with the Fraunhofer IPT and the IFU cybernetic lab of RWTH Aachen. It also involved Belgian and German end-user companies. It is now being transferred to more companies in the logistics and mechanical engineering domains. It is also available under a permissive Open Source license for all interested companies.
Continuation of Interpolative Scope and Instancing Dialogue... Development of Tools for Instances. and Testing, we are looking in v1 for the Load balancing equation... which we will state in v2 where we will make functional Keys
This document summarizes recent developments in scikit-learn, an open-source machine learning library for Python. It discusses improvements made in version 0.18, including new cross-validation objects and using randomized PCA instead of standard PCA. Upcoming improvements mentioned include adding memory caching to pipelines, a new SAGA solver for logistic regression, and quantile and local outlier factor transformers. It also discusses the scikit-learn user base of 350,000 returning users, its role as core Python infrastructure, and funding and contributions from various academic institutions that support its continued development.
SAT based planning for multiagent systemsRavi Kuril
Multi-agent Classical planning using SAT approach. This document describes the approach and discusses all the experiments and the respective results. I have considered State of the art tools for comparison purpose. Implementation code can be found on GitHub link https://github.com/ravikuril/SATbasedClassicalPlanning . For more Information contact me on ravikuril.du.or@gmail.com
1. Introduction
The Solution
Algorithm
Output
End Product
Operation Pack-Man: Shipping Estimation
Eric Bentley, Christian Bottenfield, Michael Garver, Christopher
Lindeman, Namita Matharu, Surya Padinjarekutt, Sylvia Ujwary
University of Pittsburgh
May 1, 2015
BIG Problems, Math Dept., Univ. of Pittsburgh, Spring 2015 Operation Pack-Man: Shipping Estimation
2. Introduction
The Solution
Algorithm
Output
End Product
The Client
The Problem
Redirects still-useful materials away from US landfills to
support public health programs in targeted communities
throughout the Western Hemisphere
BIG Problems, Math Dept., Univ. of Pittsburgh, Spring 2015 Operation Pack-Man: Shipping Estimation
3. Introduction
The Solution
Algorithm
Output
End Product
The Client
The Problem
Redirects still-useful materials away from US landfills to
support public health programs in targeted communities
throughout the Western Hemisphere
http://www.globallinks.com
BIG Problems, Math Dept., Univ. of Pittsburgh, Spring 2015 Operation Pack-Man: Shipping Estimation
4. Introduction
The Solution
Algorithm
Output
End Product
The Client
The Problem
The Problem
Global links needed a way to predict how many shipping
containers should be ordered for each shipment
BIG Problems, Math Dept., Univ. of Pittsburgh, Spring 2015 Operation Pack-Man: Shipping Estimation
5. Introduction
The Solution
Algorithm
Output
End Product
The Client
The Problem
The Problem
For a collection of n items {wi }, where wi = (xi , yi , zi , hi ), we
seek a mapping ϕ := {wi } → R3
BIG Problems, Math Dept., Univ. of Pittsburgh, Spring 2015 Operation Pack-Man: Shipping Estimation
6. Introduction
The Solution
Algorithm
Output
End Product
The Client
The Problem
The Problem
For a collection of n items {wi }, where wi = (xi , yi , zi , hi ), we
seek a mapping ϕ := {wi } → R3
More specifically, for a partition representing the clinics
C = {Ci } of the items {wi }, and a subspace of R3 called S,
we seek the mapping ϕc := {Ci } → S. S is given by a
rectangular prism with a corner at the origin, and restricted by
maximum values of KXC , YC , ZC .
BIG Problems, Math Dept., Univ. of Pittsburgh, Spring 2015 Operation Pack-Man: Shipping Estimation
7. Introduction
The Solution
Algorithm
Output
End Product
The Client
The Problem
The Problem
For a collection of n items {wi }, where wi = (xi , yi , zi , hi ), we
seek a mapping ϕ := {wi } → R3
More specifically, for a partition representing the clinics
C = {Ci } of the items {wi }, and a subspace of R3 called S,
we seek the mapping ϕc := {Ci } → S. S is given by a
rectangular prism with a corner at the origin, and restricted by
maximum values of KXC , YC , ZC .
K is the total number of containers needed and (XC , YC , ZC )
are the dimensions of said container.
BIG Problems, Math Dept., Univ. of Pittsburgh, Spring 2015 Operation Pack-Man: Shipping Estimation
8. Introduction
The Solution
Algorithm
Output
End Product
Stochastic Rotation
Machine Learning
Statistics
The Solution
An algorithm that takes inventory data and packs each object
into a truck
BIG Problems, Math Dept., Univ. of Pittsburgh, Spring 2015 Operation Pack-Man: Shipping Estimation
9. Introduction
The Solution
Algorithm
Output
End Product
Stochastic Rotation
Machine Learning
Statistics
The Solution
An algorithm that takes inventory data and packs each object
into a truck
Inputs:
- Dimensions of each object
BIG Problems, Math Dept., Univ. of Pittsburgh, Spring 2015 Operation Pack-Man: Shipping Estimation
10. Introduction
The Solution
Algorithm
Output
End Product
Stochastic Rotation
Machine Learning
Statistics
The Solution
An algorithm that takes inventory data and packs each object
into a truck
Inputs:
- Dimensions of each object
- Height priority of each object
BIG Problems, Math Dept., Univ. of Pittsburgh, Spring 2015 Operation Pack-Man: Shipping Estimation
11. Introduction
The Solution
Algorithm
Output
End Product
Stochastic Rotation
Machine Learning
Statistics
Stochastic Rotation
We stochastically turn and reorder items
BIG Problems, Math Dept., Univ. of Pittsburgh, Spring 2015 Operation Pack-Man: Shipping Estimation
12. Introduction
The Solution
Algorithm
Output
End Product
Stochastic Rotation
Machine Learning
Statistics
Stochastic Rotation
We stochastically turn and reorder items
For wi = (xi , yi , zi , hi ), generate βi ∈ (0, 1)and αi ∈ (0, 1)
BIG Problems, Math Dept., Univ. of Pittsburgh, Spring 2015 Operation Pack-Man: Shipping Estimation
13. Introduction
The Solution
Algorithm
Output
End Product
Stochastic Rotation
Machine Learning
Statistics
Stochastic Rotation
We stochastically turn and reorder items
For wi = (xi , yi , zi , hi ), generate βi ∈ (0, 1)and αi ∈ (0, 1)
If βi < αi then wi = (yi , xi , zi , hi )
BIG Problems, Math Dept., Univ. of Pittsburgh, Spring 2015 Operation Pack-Man: Shipping Estimation
14. Introduction
The Solution
Algorithm
Output
End Product
Stochastic Rotation
Machine Learning
Statistics
Stochastic Rotation
We stochastically turn and reorder items
For wi = (xi , yi , zi , hi ), generate βi ∈ (0, 1)and αi ∈ (0, 1)
If βi < αi then wi = (yi , xi , zi , hi )
We leverage the Central Limit Theorem to converge upon the
true mean of containers needed.
BIG Problems, Math Dept., Univ. of Pittsburgh, Spring 2015 Operation Pack-Man: Shipping Estimation
15. Introduction
The Solution
Algorithm
Output
End Product
Stochastic Rotation
Machine Learning
Statistics
Machine Learning
With our machine learning component, we approximate the
real-world solution more closely each time
BIG Problems, Math Dept., Univ. of Pittsburgh, Spring 2015 Operation Pack-Man: Shipping Estimation
16. Introduction
The Solution
Algorithm
Output
End Product
Stochastic Rotation
Machine Learning
Statistics
Machine Learning
With our machine learning component, we approximate the
real-world solution more closely each time
We adjust the dimensions based off of a formula derived by:
Ti = ˆTi + N(ˆξ − 1)3
BIG Problems, Math Dept., Univ. of Pittsburgh, Spring 2015 Operation Pack-Man: Shipping Estimation
17. Introduction
The Solution
Algorithm
Output
End Product
Stochastic Rotation
Machine Learning
Statistics
Machine Learning
With our machine learning component, we approximate the
real-world solution more closely each time
We adjust the dimensions based off of a formula derived by:
Ti = ˆTi + N(ˆξ − 1)3
We then average all past error factors to derive a current error
estimate.
BIG Problems, Math Dept., Univ. of Pittsburgh, Spring 2015 Operation Pack-Man: Shipping Estimation
18. Introduction
The Solution
Algorithm
Output
End Product
Stochastic Rotation
Machine Learning
Statistics
Statistics
Algorithm will generate and output estimates of the mean,
standard deviation, and error factor ξ for the number of
necessary containers.
BIG Problems, Math Dept., Univ. of Pittsburgh, Spring 2015 Operation Pack-Man: Shipping Estimation
21. Introduction
The Solution
Algorithm
Output
End Product
Pseudocode
Algorithm
Algorithm
We first consider the y direction.
If space exists in this direction, we fill the given space.
Otherwise, we move on to and repeat for the z direction and
then the x direction.
BIG Problems, Math Dept., Univ. of Pittsburgh, Spring 2015 Operation Pack-Man: Shipping Estimation
22. Introduction
The Solution
Algorithm
Output
End Product
Pseudocode
Algorithm
Algorithm
We first consider the y direction.
If space exists in this direction, we fill the given space.
Otherwise, we move on to and repeat for the z direction and
then the x direction.
In this way we are able to fit items on top of one another
before taking up more floor space
BIG Problems, Math Dept., Univ. of Pittsburgh, Spring 2015 Operation Pack-Man: Shipping Estimation
23. Introduction
The Solution
Algorithm
Output
End Product
Efficiency
Definition (Packing Efficiency)
Packing efficiency is the ratio of the sum of all item volumes and
the volume of all containers estimated as needed.
i (xi yi zi )
KXcYcZc
BIG Problems, Math Dept., Univ. of Pittsburgh, Spring 2015 Operation Pack-Man: Shipping Estimation
24. Introduction
The Solution
Algorithm
Output
End Product
Efficiency
Definition (Packing Efficiency)
Packing efficiency is the ratio of the sum of all item volumes and
the volume of all containers estimated as needed.
i (xi yi zi )
KXcYcZc
This number does not represent how close we have come to
any goal, as a one item overflow into a new container could
cause a drastic reduction in the calculation.
BIG Problems, Math Dept., Univ. of Pittsburgh, Spring 2015 Operation Pack-Man: Shipping Estimation
26. Introduction
The Solution
Algorithm
Output
End Product
Summary
Reception
Versatility
By interfacing with Excel, this solution can be adopted by
numerous businesses.
Since the container size is user-defined, any form of transport
can be used, whether it be by ship, air, truck, etc.
BIG Problems, Math Dept., Univ. of Pittsburgh, Spring 2015 Operation Pack-Man: Shipping Estimation
27. Introduction
The Solution
Algorithm
Output
End Product
Summary
Reception
Versatility
By interfacing with Excel, this solution can be adopted by
numerous businesses.
Since the container size is user-defined, any form of transport
can be used, whether it be by ship, air, truck, etc.
This program could benefit nearly anyone involved in
transporting and packing of goods.
BIG Problems, Math Dept., Univ. of Pittsburgh, Spring 2015 Operation Pack-Man: Shipping Estimation
28. Introduction
The Solution
Algorithm
Output
End Product
Summary
Reception
Implementation and Effects
Having the ability to determine, with certainty, the number of
shipping containers required to ship the requested supplies,
the user can know, in advance, the shipping costs that will be
associated with the order.
BIG Problems, Math Dept., Univ. of Pittsburgh, Spring 2015 Operation Pack-Man: Shipping Estimation
29. Introduction
The Solution
Algorithm
Output
End Product
Summary
Reception
Implementation and Effects
Having the ability to determine, with certainty, the number of
shipping containers required to ship the requested supplies,
the user can know, in advance, the shipping costs that will be
associated with the order.
This knowledge allows the user to accurately convey a proper
valuation of their shipment and bill the recipient properly.
BIG Problems, Math Dept., Univ. of Pittsburgh, Spring 2015 Operation Pack-Man: Shipping Estimation
30. Introduction
The Solution
Algorithm
Output
End Product
Summary
Reception
Implementation and Effects
Having the ability to determine, with certainty, the number of
shipping containers required to ship the requested supplies,
the user can know, in advance, the shipping costs that will be
associated with the order.
This knowledge allows the user to accurately convey a proper
valuation of their shipment and bill the recipient properly.
The recipient may then opt to adjust the size of their order,
either to fill a nearly full container, or to empty a barely-filled
one.
BIG Problems, Math Dept., Univ. of Pittsburgh, Spring 2015 Operation Pack-Man: Shipping Estimation
36. Introduction
The Solution
Algorithm
Output
End Product
Summary
Reception
Learning Outcomes
Throughout this project we have learned
Creation of Excel Macros
MatLab Programming
Machine Learning
Stochastic Processes
Heuristic Algorithm Development
BIG Problems, Math Dept., Univ. of Pittsburgh, Spring 2015 Operation Pack-Man: Shipping Estimation