Optimal solutions versus ‘good’ solutions: An analysis of heuristic decision making

The portion of business problems which lend themselves to solution by existing optimization techniques are a minor subset of the business problems which can be defined and attacked by logical devices. Heuristics simplify the environment of the decision maker by permitting him to make decisions quickly without considering the innumerable ways that each decision could be made. This paper compares the heuristic programming approach to more traditional operations research techniques of decision making.     

A user-focused model for the utilization of evaluation

It is proposed that the present low level of utilization of evaluation findings is traceable in part to their failure to address directly the information needs of a clearly specified decision maker. An alternative model proceeding from such specification is proposed here, with evaluation closely interwoven with the on-going innovation process. The model suggests a number of implications for the organizational role of evalutors, for the design of evaluations, and for directions for methodological development. In particular, we suggest that evaluation designs be assessed against possible “threats to utility” as well as against the traditional “threads to validity.”     

Can 401(k) accumulations generate significant income for future retirees?

This Issue Brief develops a model that project the proportion of an individual's preretirement income that might be replaced by 401(k) plan accumulations at retirement, under several different projected scenarios. The 401(k) participant behaviors in the model are based on the year-end 2000 database collected by the Employee Benefit Research Institute (EBRI) and the Investment Company Institute (ICI) in their collaborative effort known as the EBRI/ICI Participant-Directed Retirement Plan Data Collection Project. The most significant factor affecting projected replacement rates at retirement is having access to a 401(k) plan. Projected replacement rates from 401(k) accumulations at retirement are reduced significantly when participants are not offered a 401(k) plan in all portions of their careers. Most 401(k) participants tend to have contributions in any given year. Thus, projecting that participants always have contributions (their own and/or employer contributions) every year raises projected replacement rates, but not by much compared with the importance of being offered a plan to begin with. The model simulations show that participant activities such as taking loans, taking preretirement withdrawals, or cashing out account balances at job change reduce projected 401(k) accumulations and thus replacement rates at age 65. Because loans are forecast to be paid back to the account in full, their effect on replacement rates at retirement in the model is the smallest. Even if equity returns in the future are projected to replicate the worst 50-year segment in the Standard & Poor's (S&P) 500 history (1929 to 1978), 401(k) accumulations are still projected to replace significant proportions of projected preretirement income. Another projection scenario forecasts participants experiencing a simulated three-year bear market (negative equity returns) either early in their careers, near the middle of their careers, or at the end of their careers. Forecasts of the effects of bear markets on 401(k) balances show that a bear market in equities is projected to have the largest effect the closer it occurs to age 65 (retirement), even though older participants typically have diversified their portfolios away from equities. A three-year bear market for those early in their careers would reduce median replacement rates from 401(k) accumulations by an estimated 2.9-3.7 percentage points, compared with 13.4-17.7 percentage points for those immediately before retirement. Similarly, a simulated three-year bull market (positive equity returns) is projected to have a larger positive effect on projected account balances and replacement rates the closer it occurs to retirement.     

Safety Risk Analysis of an Innovative Environmental Technology

The authors describe a decision and risk analysis performed for the cleanup of a large Department of Energy mixed-waste subsurface disposal area governed by the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA). In a previous study, the authors worked with the site decision makers, state regulators, and U.S. Environmental Protection Agency regional regulators to develop a CERCLA-based multiobjective decision analysis value model and used the model to perform a screening analysis of 28 remedial alternatives. The analysis results identified an innovative technology, in situ vitrification, with high effectiveness versus cost. Since this technology had not been used on this scale before, the major uncertainties were contaminant migration and pressure buildup. Pressure buildup was a safety concern due to the potential risks to worker safety. With the help of environmental technology experts remedial alternative changes were identified to mitigate the concerns about contaminant migration and pressure buildup. The analysis results showed that the probability of an event with a risk to worker safety had been significantly reduced. Based on these results, site decision makers have refocused their test program to examine in situ vitrification and have continued the use of the CERCLA-based decision analysis methodology to analyze remedial alternatives.     

The Effects of a Stimulus-Stimulus Pairing Procedure on the Vocal Behavior of Children Diagnosed with Autism

The Analysis of Verbal Behavior - Recent research suggests that the sound produced by a child’s vocalization can become a conditioned reinforcer via the temporal pairing of an...     

The Abative Effect: A New Term to Describe the Action of Antecedents that Reduce Operant Responding

The Analysis of Verbal Behavior - Behavior-analytic terminology concerning the so-called inhibitory effect of operant antecedents lacks precision. The present paper describes the problem with...     

Woody vegetation and canopy fragmentation along a forest-to-urban gradient

To identify patterns that can be used to predict vegetation and landscape characteristics in urban environments, we surveyed the species composition and size of woody plants, as well as the landscape structure of forest canopies, along a forest-to-urban gradient near Oxford, Ohio, USA. The gradient included six sites of increasingly urban land-use: a preserve, a recreational area, a golf course, a residential subdivision, apartment complexes, and a business district. We recorded species identity and stem diameter for all woody plants greater than 3 cm diameter at breast height (DBH) to examine the distribution of individual species as well as overall community composition. We used digitized aerial photographs to compare the spatial characteristics of the forest canopy at each site. We found predictable patterns in species diversity (Shannon index), spatial heterogeneity in species composition (mean percent dissimilarity), and all measures of patch fragmentation (canopy cover and patch number and size). There were clear differences in tree density and total basal area between forested sites and developed sites, but there was little resolution among developed sites. Species richness and average DBH showed no clear pattern, suggesting that landscaping preference largely determined these values. We present a modified version of an intermediate heterogeneity model that can be used to predict diversity patterns in urban areas. We discuss probable mechanisms that led to these patterns and the potential implications for animal communities in urban environments.     

The dangers posed by saddle points, and other problems, when using central composite designs

This paper discusses two problems, which can occur when using central composite designs (CCDs), that are not generally covered in the literature but can lead to wrong decisions-and therefore incorrect models-if they are ignored. Most industrialbased experimental designs are sequential. This usually involves running as few initial tests as possible, while getting enough information as is needed to provide a reasonable approximation to reality (the screening stage). The CCD design strategy generally requires the running of a full or fractional factorial design (the cube or hypercube) with one or more additional centre points. The cube is augmented, if deemed necessary, by additional experiments known as star-points. The major problems highlighted here concern the decision to run the star points or not. If the difference between the average response at the centre of the design and the average of the cube results is significant, there is probably a need for one or more quadratic terms in the predictive model. If not, then a simpler model that includes only main effects and interactions is usually considered sufficient. This test for 'curvature' in a main effect will often fail if the design space contains or surrounds a saddle-point. Such a point may disguise the need for a quadratic term. This paper describes the occurrence of a real saddle-point from an industrial project and how this was overcome. The second problem occurs because the cube and star point portions of a CCD are sometimes run as orthogonal blocks. Indeed, theory would suggest that this is the correct procedure. However in the industrial context, where minimizing the total number of tests is at a premium, this can lead to designs with star points a long way from the cube. In such a situation, were the curvature test to be found non-significant, we could end with a model that predicted well within the cube portion of the design space but that would be unreliable in the balance of the total area of investigation. The paper discusses just such a design, one that disguised the real need for a quadratic term.     

Estimating Increment-Decrement Life Tables with Multiple Covariates from Panel Data: The Case of Active Life Expectancy*

A fundamental limitation of current multistate life table methodology-evident in recent estimates of active life expectancy for the elderly-is the inability to estimate tables from data on small longitudinal panels in the presence of multiple covariates (such as sex, race, and socioeconomic status). This paper presents an approach to such an estimation based on an isomorphism between the structure of the stochastic model underlying a conventional specification of the increment-decrement life table and that of Markov panel regression models for simple state spaces. We argue that Markov panel regression procedures can be used to provide smoothed or graduated group-specific estimates of transition probabilities that are more stable across short age intervals than those computed directly from sample data. We then join these estimates with increment-decrement life table methods to compute group-specific total, active, and dependent life expectancy estimates. To illustrate the methods, we describe an empirical application to the estimation of such life expectancies specific to sex, race, and education (years of school completed) for a longitudinal panel of elderly persons. We find that education extends both total life expectancy and active life expectancy. Education thus may serve as a powerful social protective mechanism delaying the onset of health problems at older ages.     

Population control and politics

At all times and places population and resources must be kept in balance with each other if a high, or at least acceptable, quality of life is to be maintained. Evidence is presented that thinkers and men and women of affairs have been aware of these necessities from time immemorial and that the ancients invented most of the basic concepts of modern population ecology, at leastin embryo. This task is a problem of social control, the central process in all societies. If society's controllers and their institutions are aware, intelligent, flexible, and suitably motivated, there is no reason why most problems of population, development, and pollution should not be solved and then kept permanently in check. This must involve politics, there is no way out. As Aristotle pointed out, "Man is a political animal," and it is only through politicking that we can set up, maintain, change, and generally operate our social control systems. Deeply dysfunctional attitudes towards freedom and control—especially population control—are described and analysedvia a ten-stage argument, and helpful proposals put forward. The function which scientists in general and social scientists in particular can perform in helping human-kind to solve its complex and manifold problems is heavily stressed. Scientists in general should not become politicians, but politicians in general should become numerate and "ecolate"—to quote Garrett Hardin, and start to think more deeply and longer-term. To focus practical thinking the concepts "survival-theory" and the "principle of minimum innovation" are set forth and a final plea made for the necessity and therefore moral legitimacy of controlling numbers as well as resources. If population control seems slow in returning—it used to be the norm—recent events in Eastern Europe give good grounds for optimism about the fluidity of both human values and political institutions.