Vahid Lotfi

An aspiration-level interactive model for multiple criteria decision making

Abstract A simple, eclectic approach for solving discrete alternative multiple criteria decision problems is presented. It is based on the concept of the level of aspiration, and draws on ideas of various researchers. It assumes that the user has a set of alternatives with each alternative having a score on each of a number of objectives or measures of performance. The user determines his levels of aspiration for different objectives. He is then provided with considerable feedback as to the degree of feasibility of each level of aspiration as well as the degree of feasibility with respect to all levels of aspiration as a whole. The closest nondominated solution to the solution specified by the levels of aspiration is provided. The proposed method is easy to use and easy to understand and has been implemented on a personal computer (an IBM PC or compatible with 512K RAM). We describe an experimental application in which 49 students in an MBA program used the method to solve two discrete alternative multiple criteria decision problems.

Implementing flexible automation: A multiple criteria decision making approach

This paper presents a multiple objective mixed integer linear programming model for the multiperiod flexible automation selection problem. The model includes five objectives, designed to address financial aspects, flexibility, firm disruption, and group homogeneity of new modules. The proposed model incorporates several types of flexibility. An interactive multiple criteria solution procedure is used to obtain an approximate optimal solution, based on a decision makers implicit and unknown value function. The solution provides schedule for replacement of current machines, implementation of new modules, and production plans capitalizing on flexibility during transition and subsequent periods. A numerical example is provided to illustrate the approach. The optimal solution is then contrasted with that of a single-objective model.

Flexible automation investments: A problem formulation and solution procedure

In this paper, a multi-period replacement model, based on a mixed integer nonlinear programming formulation, is developed for flexible automation investments. The model takes into account the costs, benefits and effective utilization of several types of flexibility. The decision variables pertain to the selection and optimal implementation sequence for new, CNC modules, the replacement schedule for current equipment and the aggregate production plans for transition and subsequent periods in the planning horizon. The objective function maximizes the present worth of the cash flows over the planning horizon. A two-level, exact solution method is also developed, utilizing dynamic programming methodology for the higher level sub-problem and mixed integer-linear programming for the lower level sub-problem.

A Comparison of Aspiration Level Interactive Method (AIM) and Conjoint Analysis in Multiple Criteria Decision Making

The predictive validity of the two methods, Aspiration-level interactive method (AIM) and conjoint analysis, used for solving decision problems involving discrete alternatives are compared. An empirical analysis based on subjects’ preferences for a multiattribute product (buying a house) and a service (selecting an MBA program for study) indicated that consumer preferences derived from AIM may be more valid than the preferences derived from the full-profile conjoint analysis method.

A dynamic approach for outpatient scheduling

Abstract Aims: Patient no-show is a recurrent problem in medical centers and, in conjunction with cancellation of appointments, often results in loss of productivity and excessive patient time to appointment. The purpose of this study was to develop a dynamic procedure for scheduling patients within an outpatient clinic where patients are expected to have multiple appointments, such as physical therapy, occupational therapy, primary care, and dentistry. Methods: This retrospective study involved the year 2014 de-identified patient records from an outpatient clinic affiliated with a large university hospital. A number of patient characteristics, appointment data, and historical attendance records were examined to determine whether they significantly impacted patients who missed scheduled appointments (no-shows). Patient attendance behaviors over multiple appointments were examined to determine whether their no-show and cancellation patterns differed from one appointment to the next. Decision tree analysis was applied to those predictors that significantly correlated with patient attendance behavior to assess the likelihood of a patient no-show. A sample dynamic appointment scheduling procedure that utilized different overbooking strategies for different appointment numbers was then developed. Computer simulation was used to assess the effectiveness of the dynamic procedure versus two other methods consisting of randomly assigned and uniformly assigned appointments. Results: The dynamic scheduling procedure resulted in increased scheduling efficiency through overbooking but with less than 5% risk of appointment conflicts (i.e. two patients showing at the same time), equating to approximately 0.16 conflicts per clinician per day. It also increased clinic utilization by about 6.7%. It consistently outperformed the other two methods with respect to the percentage of appointment conflicts. Limitations: The study is limited with respect to potential clinician cost increase resulting from possible appointment conflicts. A second limitation is that patients experiencing appointment conflicts might not wait for treatment, resulting in potential loss of revenue. A third limitation is that the model does not take into account patient satisfaction, nor the ethics of overbooking patients. Conclusions: A dynamic appointment scheduling procedure was developed using actual patient characteristics. The procedure resulted in creation of more efficient appointment schedules thereby increasing the clinic utilization.

Reply: A dynamic approach for outpatient scheduling

We found this letter to present an interesting option for addressing the problem of “missed appointments”. The authors suggest exploring “evening clinics” as an option to mitigate the pressure on service and excessive patient waiting time. The authors report positive results from a “pilot” program in which they had used an evening out-of-hours service in a pediatric clinic for 4 weeks in 2015. The above option certainly merits further exploration in an environment where there is pressure on service providers, resulting in excessive patient waiting time. The issue at the authors’ clinic seems to be insufficient capacity, and the option of adding “out-of-hours” service results in increased clinic service capacity. In our study, the situation was quite different. There was, in fact, excess capacity, as established by the relatively low clinician utilization of only 65.1%. In addition, the patients’ missed appointments in the clinic were not caused by excessive waiting times. Rather, they were partly due to the number of days to the scheduled appointment and mostly because of the patients’ population life circumstances. We believe controlled overbooking in such an environment, with rather sub-optimal productivity, is quite justified, especially if one is able to control the risk of potential appointment conflicts, as we have demonstrated.