Raja Jayaraman

Support vector-based algorithms with weighted dynamic time warping kernel function for time series classification

In this paper, we propose support vector-based supervised learning algorithms, called multiclass support vector data description with weighted dynamic time warping kernel function (MSVDD-WDTWK) and multiclass support vector machines with weighted dynamic time warping kernel function (MSVM-WDTWK), which provides a flexible and robust kernel function for time series classification between non-aligned time series data resulting in improved accuracy. The proposed WDTW kernel function provides an optimal match between two time series data by not only allowing a non-linear mapping between two data sequences, but also considering relative significance depending on the phase difference between points on time series data. We validate the proposed approaches using extensive numerical experiments on a number of multiclass UCR time series data mining archive, and demonstrate that our proposed methods provide lower classification error rates compared with existing techniques.

A polynomial goal programming model with application to energy consumption and emissions in United Arab Emirates

Sustainable growth and development of nations requires strategic vision and plan to accommodate growing energy needs while preserving and protecting the natural environment. Studies that couple energy and environment often present conflicting objectives which requires the explicit use of quantitative models to develop appropriate solutions for policy planning and analysis. Goal programming offers an analytical framework to solve multi-objective problems that simultaneously satisfy decision makers preferences. In this paper we develop a polynomial goal programming model to study the effects of electricity consumption and greenhouse gas emissions applied to the economy of United Arab Emirates. The results provide valuable insights for improvement opportunities and quantitative justification for the required investments and efforts in implementing sustainable development plans.

A decision support tool for healthcare providers to evaluate readiness and impacts of adopting supply chain data standards

Healthcare providers are under increasing pressure to reduce waste, eliminate unnecessary costs and redundant efforts, thereby improving the quality and consistency of healthcare delivery. Lack of automation and the lack of use of global identifiers for products and locations, also known as supply chain data standards, are two critical factors that can help streamline providers operations and improve process efficiency. Despite widespread consensus among various stakeholders, healthcare providers lack a well-defined approach towards adopting and implementing data standards. Supply chain data standards can be defined as a set of product and location identifiers which are used in supply chain related processes and transactions. Healthcare providers willing to successfully adopt data standards in some or all of their operations need to invest in several process changes and technology installations or upgrades; however, they often struggle to justify returns on those investments and hence, find uncertain Return on Investment (ROI) as a critical barrier. In this article, we present a hierarchical comprehensive spreadsheet based decision support tool that helps potential healthcare providers to evaluate their readiness requirements and quantify the potential impacts of their decisions in terms of non-monetary performance measures, such as increased productivity, enhanced patient safety and reduction in errors resulting in decreased volume of transactions. This tool has undergone extensive testing with healthcare providers of different size, scope, and needs. We present numerical results showcased through practical examples in this article. The software is publicly available free of cost for download at http://cihl.uark.edu.

Optimal control with multiple human papillomavirus vaccines

A two-sex, deterministic ordinary differential equations model for human papillomavirus (HPV) is constructed and analyzed for optimal control strategies in a vaccination program administering three types of vaccines in the female population: a bivalent vaccine that targets two HPV types and provides longer duration of protection and cross-protection against some non-target types, a quadrivalent vaccine which targets an additional two HPV types, and a nonavalent vaccine which targets nine HPV types (including those covered by the quadrivalent vaccine), but with lesser type-specific efficacy. Considering constant vaccination controls, the disease-free equilibrium and the effective reproduction number Rv for the autonomous model are computed in terms of the model parameters. Local-asymptotic stability of the disease-free equilibrium is established in terms of Rv. Uncertainty and Sensitivity analyses are carried out to study the influence of various important model parameters on the HPV infection prevalence. Assuming the HPV infection prevalence in the population under the constant control, optimal control theory is used to devise optimal vaccination strategies for the associated non-autonomous model when the vaccination rates are functions of time. The impact of these strategies on the number of infected individuals and the accumulated cost is assessed and compared with the constant control case. Switch times from one vaccine combination to a different combination including the nonavalent vaccine are assessed during an optimally designed HPV immunization program.

On prorated servicing costs for two-dimensional warranties with combined repair-replacement strategies

We study a non-renewing two-dimensional warranty policy for repairable products with pro-rated cost for rectifying failures within a warranty coverage period defined by a rectangular region. Product failures in different warranty regions have different cost implications based on repair-replacement strategy employed by the manufacturer and the time and usage at the event of product failure. We obtain expressions for expected cost to the manufacturer based on two possible repair-replacement strategies which can be employed towards servicing the product in the event of failures and present the respective costs associated with them.

A Goal Programming model with satisfaction function for long-run sustainability in the United Arab Emirates

Goal Programming (GP) is a relevant and simple technique in Operations Research that helps Decision Makers in solving problems involving conflicting and competing criteria and objectives. A GP program is a distance-based model that involves the minimization of the difference between the achievement level of a certain criterion and its goal. GP models when combined with the notion of satisfaction or utility function provide practical advantage to the decision maker to include a system of preferences. In this paper, we present a GP model with satisfaction function that concurrently incorporates criteria on number of employees across various economic sectors and multiple sustainability goals on GDP growth, electricity consumption and Greenhouse Gas (GHG) emissions. The presented model offers critical inputs in planning and implementation of amenable strategies towards sustainable development. We validate the proposed model with data on vital economic sectors of the United Arab Emirates (UAE) towards achieving year 2030 goals.

Going beyond healthcare IT inter-operability in chronic disease management

Chronic care consumes the majority of a nations healthcare costs as it involves more specialists providing care over multiple visits for an extended duration of time. The role of healthcare information technology (HIT) in managing chronic care presents a significant opportunity for improvement. Interoperability has been seen as the stumbling block to HIT for achieving efficiencies in paperless patient care. Yet from a systems perspective, the integration challenge to effective delivery of healthcare appears to go beyond inter-operable HIT, especially in chronic disease management. A great deal of interaction takes place beyond the limited scope of information passed between disparate HIT systems. For example, the chronic patient plays a greater role through self-monitoring which means applications of HIT must extend beyond the premises of the various providers. This paper presents a systems approach to identify the key technologies and people involved in chronic care and more importantly, the likely interactions between them. An understanding of these system elements then become the basis for prioritizing future research and clinical interventions including the necessary scope and scale of inter-operability.

Integrating supply chain data standards in healthcare operations and Electronic Health Records

Electronic Health Records (EHRs) together with supporting information systems offer myriad possibilities for system wide improvements to data access, work flow and integrated information exchange. Parallel to these developments providers must implement supply chain based data standards to their operations in order to realize system interoperability and comprehensive data integration. Supply chain data standards are universal, unique and unambiguous identifiers for products and locations. Integrating supply chain standards in provider processes, clinical workflow, and ultimately in EHRs offers an ability to track products, serves as a basis for cost and clinical effectiveness, improves patient safety and quality outcomes. In this paper, we present the significance and benefits of adopting and transacting using supply chain data standards, improvement areas within provider operations, potential advantages of integrating supply chain standards with EHRs and barriers to healthcare provider implementation.

An Exploratory Pilot Study on Supply Chain Data Standards in a Hospital Pharmacy

Abstract Supply chain data standards provide unique and unambiguous references to products and locations, enable system level interoperability, facilitate barcode-based medication administration, and offer track-and-trace capabilities throughout the healthcare supply chain. Despite widespread consensus among industry stakeholders and many perceived benefits, healthcare providers are hesitant to implement these standards. Through an exploratory pilot study at a healthcare provider site focused on pharmaceutical operations, we identify the key challenges in implementing data standards in a pharmaceutical setting. In doing so, we document the required technology functionality, clinical and financial workflows, and information systems, as well as articulate the challenges in implementing data standards. The lessons learned from the study offer valuable insights for supply chain executives, engineering managers, and pharmacy directors to plan and implement GS1 standards in their facilities.