Mary Tom

Family System: A Reference Model for Developing Home Automation Applications

This research is a pioneer attempt to establish theoretical foundation and a framework for technical developments in home automation. This study suggests the need for an International Technology Roadmap for Home Automation (ITRHA) to formulate a need based vision of typically 10 to 15 years and stresses the need for HA to cater for the specific requirements of family home users. The concept of family life cycle is used to understand the temporal needs of family. In order to formally describe a coherent set of family processes, their relationships, and interaction with external elements, a reference model named family system is established that identifies external entities, 7 major processes, and 7 subsystems to manage smooth functioning of these processes by allocating resources and interfacing between processes and external entities. The family system reference model can be used to conduct detailed requirement analysis and thus identify potential HA products and services meeting realistic user needs.

Pixel characteristics based feature extraction approach for roadside object detection

Classification of roadside objects is very important task in identifying fire risk regions, analysing roadside conditions and improving roadside safety. This paper introduces a novel and effective way to detect soil, grass, road and tree from roadside images thus giving a better decision-making system for analysing roadside video data. A new feature extraction approach is proposed to detect and classify the roadside objects. Feature set is based on colour characteristics which are obtained by analysing components of image pixels. Choosing an appropriate feature set is one of the great challenges for successful identification of roadside objects. Based on the proposed feature set and the Support Vector Machine, the detection and classification approach is implemented. The proposed approach is evaluated using the training and test data from real-world roadside video images. The results show that the proposed approach is able to accurately detect grass, soil, road and tree.

Requirement Analysis of Ubiquitous Intelligence System for Personal Diet Formulation and Maintenance

Nutrition Related Diseases (NRD) are globally widespread causing substantial economic cost for governments and individuals. Formulating and maintaining a balanced diet meeting the Recommended Dietary Intake (RDI) is complex, time consuming task. The detailed analysis conducted in this study reveals many intellectual tasks involved in meals scheduling and diet formulation. This study reports the results of requirement analysis of a Ubiquitous Intelligence System for automating the diet formulation, monitoring and maintenance. The authors propose Dietary Intelligence System (DIligenS) that applies computational intelligence and has the architectural components of knowledge base, inference engine, data base and data extraction and conversion modules in addition to the knowledge building and application interface.

Future user requirement elicitation for technology investment: A formal approach

This study addresses the lack of methods to systematically identify future market needs for strategic planning in technology investment. Uncertain and ambiguous future, hardly available user input on realistic needs, market with latent demand and evolving market increase technology investment risk. The authors presents a formal approach named Requirement Elicitation of Future Users by Systems Scenarios (REFUSS) that derives future user requirements by relating process knowledge obtained from current system model with plausible future user lifestyle related information obtained from scenarios. Scenarios developed in a cause-effect way by identifying and correlating a number of environmental factors assists strategists to learn about plausible changes and be prepared for any unexpected shift. The REFUSS fills a gap by providing a simple and applicable method for future user requirement elicitation. This can be used for formulating needs based vision thus avoiding project failures from following technology based vision.

A fuzzy multicriteria decision making model for restaurant menu ranking

Non Communicable Diseases (NCD), mainly caused by imbalanced diet, such as type 2 diabetes and cardiovascular diseases are increasing. At the same time, Food Away From Home (FAFH) consumption is on the rise. Menu items include categories and sub-categories, and each menu item has five or more critical nutrients to be evaluated such as calorie, cholesterol, saturated fat, sodium, fiber and others. Comparing and choosing healthier menu choices becomes a difficult and time consuming task. Different consumers have varying preferences depending on their health conditions and other preferences. To assist consumers in this task, this paper presents a fuzzy multicriteria decision making model for ranking menu items based on the customers chosen prioritized criteria. An effective algorithm is developed to rank menu items. An example is presented for demonstrating the applicability of the model in ranking actual menu items listed with necessary nutritional information.

Computational intelligence using Fuzzy Multicriteria Decision Making for DIligenS: Dietary Intelligence System

Treatment of Nutrition Related Diseases (NRD) mostly arising from imbalanced diet has an associated high economic cost globally. Planning meals and formulating a balanced diet are time consuming, complex routine decision making involving constraint satisfaction and meeting multi-part objective of the Recommended Dietary Intake (RDI). The authors have proposed DIligenS - Dietary Intelligence System to provide ubiquitous intelligence for automating the diet formulation and monitoring. DIligenS has a seven layered architecture consisting of knowledge base and inference engine, and necessary database. The meals scheduling, and diet formulation including consumption of food away from home is to be carried out within many constraints of budget, time, health, and individual preferences. This study investigates the applicability of Fuzzy Multicriteria Decision Making (FCDM) following the Prioritised Fuzzy Constraint Satisfaction Approach to computerise the inherently imprecise and vague user preferences for meals selection.

A fuzzy multi-criteria decision making model for menu engineering

Menu remains a key element in influencing the success of restaurants which is very dynamic and highly competitive with a high failure rate within the first three years of operation. Menu engineering refers to the specific techniques used to evaluate performance of individual menu items leading to strategic decision. Many influencing elements are consolidated to two elements of popularity index and contribution margin providing four different combinations that can be used to choose decision options. Decision makers carry out menu engineering against manually set target values which are imprecise and choice of strategic options becomes erroneous and tedious. As a step toward providing a more powerful decision making tool this study presents a fuzzy multi-criteria decision making model to choose strategy decision options that extends the set of combinations from four to nine. The model uses trapezoidal fuzzy numbers for normalization and linguistic variables for fuzzification. The applicability of the model is tested using thirty menu items in four categories.

Optimized daily diet composition for a nutritionally balanced diet: An application of Fuzzy Multiple Objective Linear Programming

Nutrition related diseases such as Cardiovascular Diseases, type 2 diabetes, and certain types of cancers are widely prevalent, and increasing in both developed and developing countries. This escalates the need to consistently follow a nutritionally balanced diet. Formulation of nutritionally balanced Meals is a complex task, which requires considerable effort, time, and intellectual capability for evaluation of Menu items and analysis of a range of nutrient values. As a step towards automating this task, this interdisciplinary study presents a fuzzy optimization model applying Fuzzy Multi-objective Linear Programming (FMOLP) to compose a daily diet consisting of nutritionally balanced Meals which limit nutrients such as cholesterol, sugar, saturated fat, and sodium. An effective algorithm is developed to create a pare to optimal solution set of optimized daily diet choices. The model is tested using one hundred Menu items, for demonstrating the applicability of the model in composing a nutritionally balanced diet.

A multicriteria analysis approach for evaluating the sustainability performance of hydropower projects under hesitant fuzzy environment

Hydropower is recognized as one of the most important clean and renewable energy choices which fuel the overall societal development by contributing a notable proportion of sustainable energy to the whole global power generation. Although much attention has been paid to the development of hydropower plants, few of them attempted to study the sustainability performance of the available hydropower project alternatives for achieving an environmentally sustainable and socially equitable manner. This paper develops a multicriteria analysis approach for evaluating the performance of hydropower projects under hesitant fuzzy environment. Hesitant fuzzy set is used to deal with the conditions where the decision maker hesitates among several values to assess the alternatives. An effective algorithm is developed for generating a performance index for every hydropower project alternative across all evaluation criteria. An example case is presented to demonstrate the application of the developed approach for dealing with a hydropower evaluation and selection problem.