ITECH7413- SUPPLY CHAIN MANAGEMENT TEAM ASSIGNMENTWEIGHTING: 25% (15% for the report and 10% for the presentation) TASK and Due WeekGroup Presentation (Due Date - Week 10 – Allocated...

Running Head: Research report        1                                                
Research Report            
GROUP REPORT
HOW MACHINE LEARNING CAN TRANSFORM THE SUPPLY CHAIN
Executive summary
This report discusses about significance of machine learning in transforming the supply chain. This report shall help to understand supply chain and machine learning as individual terms and also in relation to each other. Objectives of making use of machine learning in supply chain have been discussed here. A detailed explanation has been given to understand how machine learning can transform supply chain. Methods to do the same have been discussed along with the benefits and challenges. The assistance which machine learning will provide is also elaborated in this research report.
Table of contents
Executive summary    2
Introduction    4
Objective    5
Machine learning in supply chain operations    6
Forecasting demand    6
Supply forecasting    6
Data analytics    6
Price planning    7
Inventory planning    7
Recommendations for shipment    7
Stock analytics    7
Exception analytics    8
Analytics at component level    8
Production planning    8
Supply chain operation optimization with machine learning    8
Benefits of using machine learning in supply chain    9
Challenges of using machine learning in supply chain    11
Stages in implementation of machine learning    11
Practical implementation of machine learning using IOT    12
Artificial intelligence in machine learning    13
Neural networks in machine learning    13
Machine learning assistance in strategic decision making    14
Machine learning in real time decision making in supply chain    14
Conclusion    15
References    16
Appendices    20
Introduction
Machine learning is becoming a most looked for technology to transform supply chain by solving challenges of supply chain such as time, cost and resources. Machine learning makes use of several algorithms to find anomalies in huge data sets. They also help to study the existing pattern and gain predictive insights into large data sets. This will help to make any future forecasts more accurately. Machine learning will not only help to control the costs but will also help to manage the schedule. To understand the significance of machine learning in supply chain it is important to understand the two terms individually. Supply chain can be defined as network of processes undertaken by a company along with its suppliers so that the company can produce and distribute its products to the buyers. It can also be defined as series of steps taken to carry a product or service form its initial state to the customers. Machine learning can be defined as an application of AI that is automated to gain insights and improve the system based on the gained insights form the existing patterns, data or past experiences. After learning these two terms it is now easy to understand the importance of machine learning in transforming the supply chain. It could be inferred that evaluates the effects of emerging technologies and thus helps to enhance the performance of supply chain systems. To carry out improvement in the development of supply chain it is essential to incorporate machine learning systems.
Objective
The primary objective behind using machine learning for improvement or transformation of supply chain is to reduce the operational inefficiencies which may result in increasing costs, losses in revenue and poor customer satisfaction. Analytical tools can be used to make well planned and informed strategic decisions to avoid any kind of loss for the organisation. Machine learning will be able to discover patterns in supply chain so that most influential factors can be determined which will facilitate in the success of supply chain. Kartal, Oztekin, Gunasekara and Cebi (2016) have mentioned in their work that in the modern world machine learning algorithms are used to discover the patterns and gain an insight into meaningful and useful information form a large amount of data available. Another important objective behind the deployment of machine learning is to make use of real time data analytics which will help to predict the challenges well in advance. The purpose of incorporating machine learning algorithms is to facilitate the interchange of data between the company and its suppliers and between the company and its customers by electronic means. Machine learning algorithms can be used to know the key factors which a have a great influence on inventory levels, forecast of demands, order to cash and procure to pay. With machine learning there is no need of manual intervention to carry the analysis of the data available in large volumes. These algorithms also make electronic transactions between organisation, suppliers and customers much easy. The aim is to use to technology to increase efficiency and productivity of the supply chain and to make it free of errors as much as possible.
Machine learning in supply chain operations
Machine learning can completely transform the supply chain by focusing on the following activities or processes of the supply chain:
Forecasting demand: machine learning can assist in forecasting demand for the future accurately without many errors. As Carbonneau, Laframboise and Vahidov (2008) have mentioned in their work that machine learning can provide collaborative forecasting and replenishment that will allow a company and its suppliers to coordinate decisions by exchanging data, information, complex models and strategies that will support decision making for the future forecast. Machine learning algorithms can support various forecasting techniques such as multiple linear regression, naïve forecast, trends in the industry etc.
Supply forecasting: accurate predictions of future supply can be made by structuring the purchase orders and the bill of material. Machine learning algorithms can predict the pattern of supplier lead times and commitments and thus can be utilized to facilitate supply forecasting. As suggested by Ulbricht, Fischer, Lehner and Donker (2013) machine learning algorithms can extract meaningful information from the data provided by classic applications such as time series analysis method which will help to improve the process of supply chain forecast.
Data analytics: machine learning can be applied to voluminous data to clean it so that a more understandable master data can be derived. Analytics can be applied to data of suppliers as well as data of shipments to gain an insight into huge data available. As mentioned by Rozados and Tjahjono (2014) in their work that big data analytics has transformed the processes of supply chain leading to improvement in operational efficiency and business transformation.
Price planning: machine learning algorithms can observe the patterns demand trends and lifecycle of the product based on the prices automatically without any manual intervention. This will help the firm to make strategic decisions of increasing or decreasing the price of their product or services based on the insight gained through machine learning algorithms. Kovalchuk and Fasli (2008) have mentioned in their work that machine learning can be used to predict customer offer prices by following two way approaches i.e. by predicting the bidding prices of the competitors and by predicting lowest order prices of the products.
Inventory planning: application of machine learning algorithms to inventory planning can be beneficial for supply chain these algorithms are capable of automatically raising purchase orders based on existing shortages or future demands. Alicke, Rexhausen and Seyfert (2017) have mentioned in their work that using machine learning in inventory planning can help to improve the profile of inventory and will consequently lead to lower costs and better levels of service.
Recommendations for shipment: based on the buying behavior of the customers in the past it can automatically recommend the products which are in access and can suggest strategies to clear the inventory of such products. Kasarda (2017) has mentioned that machine learning can be an essential element of...