ANLY482 AY2017-18T2 Group10 Project Overview

HOME

ABOUT US

PROJECT OVERVIEW

ANALYSIS & FINDINGS

PROJECT MANAGEMENT

BACK TO MAIN ANLY482

Overview

Data

Methodology

Sponsor Background

Our sponsor is a food-service organisation that owns and manages various renowned restaurant brands. The group was founded in the 1980s in Singapore, starting with a fine-dining Korean restaurant for its first brand. Since then, it has evolved into different concepts, with further overseas expansion. Under these concepts, the group has a total of 13 outlets in Singapore.

Project Overview

For the scope of our practicum, we will be focusing only on the Singapore operations of one particular concept. Currently, our sponsor company is not able to accurately determine the amount of ingredients to order for their chain of restaurants. It is often based on guesswork and gut feeling which has often led to excessive holding costs as well as food wastage and in some cases, shortage of ingredients. This is not ideal as it may lead to various issues such as cost due to the non-freshness of the product, throwing of food due to expiry, not enough space in storage, the inability to satisfy customer demand, etc.

We will be utilising inventory, Programmable Logic Unit (PLU) and sales data to forecast demand of customers and also for each ingredient. As there is a long list of ingredients, we will be focusing on the more popular ingredients first. After predicting the likely amount to order for each ingredient, we will develop a user-friendly application which enables the staff working in each outlet to easily view the predicted amount to order for each ingredient.

For our research paper, our focus will instead be on customer count forecasting where we will compare between two time-forecasting methods (i) exponential smoothing and (ii) autoregressive integrated moving average (ARIMA)

Project Objectives

Our first objective is to develop a model which helps individual outlets predict the amount of ingredients needed each month as well as the customer count forecast for each week. With this model, the staff can more accurately gauge the optimal inventory quantity and order quantity and not need to rely on gut feeling from previous experience, as is the existing practice. While the customer count forecast helps outlet managers gauge the number of staff required and in planning of staff schedule. In summary, our business goal is to ensure accurate and optimal orders to fulfil storage, optimising storage space for each individual outlet.

Motivations

There have been several areas of concern raised by our sponsor pertaining to inventory management. In particular, achieving a balance between not holding excessive inventory and ensuring that there are sufficient items for customers at all times. Additionally, ingredients should not be kept for too long as this has a negative impact on its freshness and may result in the disposal of it. By predicting the amount to order, we would also reduce wastage and be more environmentally friendly. Lastly, to reduce cost attributed to food wastage, as well as reducing the time taken by the staff to determine the order quantity for the various outlets.

As for forecasting customer count, the quantity of food to provide and order is highly dependent on the number of expected customers, and thus being able to accurately forecast the number of customers could have wide-ranging implications for a business such as customer satisfaction, costs (both ordering and holding) and staff scheduling.

From our group’s perspective, we hope to learn various soft skills, such as requirement gathering, asking the right questions and stakeholder management. In addition, we believe that through this, we will be able to apply what we learnt about analytics into a real world project from start to finish. Lastly, we want to cultivate an independent learning attitude in our team and to be open to new methods and ideas.

Stakeholders

ABC Company
Mdm Meenakshi Gopalakrishan, SIS Instructor, Project Supervisor