The development and adoption of artificial intelligence (AI) and robotic technologies in the foodservice industry has expanded dramatically. The economic benefits of such adoption are likely to be similar to those experienced by other sectors, such as manufacturing. However, unlike many other sectors, the appeal of restaurants involves consumer perceptions of those making the product. In particular, we argue that—especially in craft food contexts—consumers expect food to be prepared “with love.” As robots are intuitively incapable of doing so, restaurateurs face a conundrum: how to take advantage of the economic benefits of robotic chefs, while maintaining consumer perceptions that meals are prepared with love? We test a series of potential interventions aimed at overcoming the gap in such perceptions and find that AI-enabled (i.e., chat-based) relationship-building between robot chefs and restaurant patrons is the most effective option. In fact, our relationship-building intervention fully closes the gap between preferences for human chefs, relative to robotic chefs. Additional managerial and theoretical implications are discussed.

The purpose of this conceptual article is to document major threats to online survey data quality through the lens of current practices in the marketing research industry. The Fair Framework categorizes threats ranging from respondent misbehavior to criminal activity in the form of survey fraud. Real examples provide evidence of threats by individual humans as well as serious weaknesses undermining the research supply chain infrastructure, especially regarding sampling. This paper offers academic and practitioner researchers detailed knowledge and actionable insights to effectively combat data quality threats and support the development of robust, reliable, and ethical marketing research practices.

Misleading information pervades marketing communications, and is a long-standing issue in business ethics. Regulators place a heavy burden on consumers to detect misleading information, and a number of studies have shown training can improve their ability to do so. However, the possible side effects have largely gone unexamined. We provide evidence for one such side-effect, whereby training consumers to detect a specific tactic (illegitimate endorsers), leaves them more vulnerable to a second tactic included in the same ad (a restrictive qualifying footnote), relative to untrained controls. We update standard notions of persuasion knowledge using a goal systems approach that allows for multiple vigilance goals to explain such side-effects in terms of goal shielding , which is a generally adaptive process by which activation and/or fulfillment of a low-level goal inhibits alternative detection goals. Furthermore, the same goal systems logic is used to develop a more general form of training that activates a higher-level goal (general skepticism). This more general training improved detection of a broader set of tactics without the negative goal shielding side effect.

The authors ask whether individuals tasked with persuading others have distinct and important concerns regarding their occupational stress and well-being. The authors argue that a well-known model from the marketing literature – the persuasion knowledge model (PKM; Friestad & Wright, 1994) – illuminates a number of issues for future study. The authors further argue for a number of extensions to the PKM to account for the persuasion agent’s side of the interaction. Next, the authors consider potential stressors that are distinctive to the persuasion encounter, as well as the strategies that persuasion agents engage to cope. This discussion reveals a number of potential negative consequences for the agents themselves, as well as their employing firms and customers. Finally, the authors present some thoughts on what persuasion agents, their managers, and external regulators can do to mitigate these negative consequences.

It is widely accepted, and demonstrated in the marketing literature, that negative online word of mouth (NOWOM) has a negative impact on brands. The present research, however, finds the opposite effect among individuals who feel a close personal connection to the brand—a group that often contains the brand’s best customers. A series of three studies show that, when self–brand connection (SBC) is high, consumers process NOWOM defensively—a process that actually increases their behavioral intentions toward the brand. Study 1 demonstrates this effect using an experimental manipulation of SBC related to clothing brands, and provides process evidence by analyzing coded thought listings. Study 2 provides convergent evidence by measuring SBC associated with smartphones, and followup analyses show that as SBC increases, the otherwise negative effect of NOWOM steadily transforms to become significantly positive. Study 3 replicates these results using a combination of a national survey conducted by J.D. Power investigating hotel stays and data drawn from TripAdvisor. Results of all three studies, set in product categories with varying levels of identity relevance, support the positive effects of NOWOM for high-SBC customers and have implications for both managers and researchers.

E-commerce offers retailers the opportunity to attract new customers online; however, consumer distrust toward unfamiliar retailers can seriously impede these efforts. Construal Level Theory suggests that such distrust can be partially understood in terms of psychological distance, and that reducing psychological distance using simple website tactics should overcome distrust and encourage first-time purchases. Studies 1 and 2 show a physically distant retail store, or lack of a physical store altogether, contribute to psychological distance, distrust, and reluctance to purchase online. Studies 2 and 3 further show that website images of an office building (increased tangibility), or the owner's name and appearance (social proximity), can improve trust and purchase intentions by specifically reducing the psychological distance otherwise associated with purely virtual or physically distant retailers. Published by Elsevier Inc.

In a process the authors term just world coping, some consumers use positive beliefs concerning the general benevolence of the world as a resource to cope with marketplace threat. This belief buffers or even, ironically, enhances trust judgments in the face of threat. Three experiments and one replication show that, whereas consumers who do not hold this belief respond to decision-generated threat with distrust, trust is significantly higher for those who believe in a just world (optimistic trust effect). Process evidence shows such coping is automatically activated in response to threat but can be corrected for more normative considerations when an obvious ulterior motive is present. Finally, evidence this coping serves an ego-protective function is provided by manipulating whether consumers are directly threatened. Overall, findings are consistent with the view that belief in a just world operates as a positive illusion that allows consumers to cope with decision threat.

Few administrative units involved in wildland fire protection are islands unto themselves when it comes to wildfire activity and suppression. If not directly affected by the wildfire workload of their neighbors, they are affected by the availability of nationally shared resources impacted by wildfire activity at the regional and national scale. These external influences should be taken into account when local administrative units plan their presuppression organizations. Failure to account for the external influences of fire workload and resource availability can lead to incorrect assessment of local resource needs. Recent advances in fire planning technology funded by the National Fire Plan have made it possible to better account for these external forces. Testing this new fire planning technology on the Umatilla National Forest in eastern Oregon and Washington, we found common planning assumptions about the influences of external fire activity can lead to significant errors in estimating suppression program performance.

Consumer innovativeness and new product purchasing literatures are replete with solid yet unrelated theories that have not been considered simultaneously as part of a larger psychological framework. This oversight limits the ability of practitioners to effectively target the valuable consumer innovators market segment. In this study, an approach/avoidance framework of new product purchase intentions is discussed and empirically tested via structural equation modeling. Consumer innovativeness, self-congruence, and satisfaction play the role of approach mechanisms, while perceived risk acts as an avoidance mechanism. The authors combine a set of related yet disconnected theories, while suggesting a means of appealing to consumer innovators through a specific form of self-congruence. A sample of 741 students is employed to examine these issues. Several notable findings are highlighted, including verification of indirect relationships between the independent variables and behavioral intent. Model fit is excellent and results are consistent across the handheld devices, home entertainment, and music industries. © 2008 Wiley Periodicals, Inc.

An algorithm is described that quickly calculates minimum travel times between locations for initial response forest fire suppression units. The algorithm was developed for integration into wildfire planning simulation models to quickly identify fire suppression units with the fastest response to a fire during a simulation. While generally quite fast, network-based path minimization algorithms are not suitable for the wildfire problem since fires are often located away from existing network features, such as roads and trails. On the other hand, traditional raster-based least-cost path algorithms can take minutes, if not hours, to calculate each resource’s travel time, too slow to be effective within a simulation model. The algorithm presented in this paper enhances the raster-based technology, providing a multi-resolution approach. Our test results show the new algorithm dramatically increases the speed of calculations with only limited loss of accuracy Introduction One vexing problem confronting the design of realistic simulation models for initial attack wildfire planning is accurately estimating ground suppression resource travel times between locations. From the standpoint of initial attack simulation model building, there is more than just a need for accurate estimates of travel times; there is also a need within the simulation model to quickly make these calculations thereby maintaining reasonable simulation run times. In this paper, we describe a computationally efficient shortest-path algorithm for computing travel times within an initial attack wildfire simulation model. This algorithm, COSTPATHSQ (Q for quick), was developed to meet the needs of the Wildfire Initial Response Assessment System (WIRAS), a comprehensive wildfire initial attack planning tool (Wiitala and Wilson, this proceedings). 1 An abbreviated version of this paper was presented at the second international symposium on fire economics, planning, and policy: a global view, 19–22 April 2004, Córdoba, Spain. 2 GIS Developer, Digital Visions Enterprise Unit, Forest Service, U.S. Department of Agriculture, 333 SW 1st Ave, Portland, OR, 97204. Operations Research Analyst, Forestry Sciences Laboratory, Forest Service, U.S. Department of Agriculture, 620 SW Main Street, Suite 400, Portland, OR 97205. Resource Information Specialist, Forestry Sciences Laboratory, Pacific Southwest Research Station, U.S. Department of Agriculture, 620 SW Main Street, Suite 400, Portland, OR 97205. GIS Analyst, Digital Visions Enterprise Unit, Forest Service, U.S. Department of Agriculture, 1230 NE Third St., Suite A-262, Bend, OR 97701. USDA Forest Service Gen. Tech. Rep. PSW-GTR-xxx. xxxx. 2 Proceedings of the Second International Symposium on Fire Economics, Planning, and Policy: A Global View GENERAL TECHNICAL REPORT PSW-GTR-208 Session Poster— A Fast Method for Calculating Emergency Response Times —Hatfield, Wiitala, Wilson, and Levy WIRAS simulates the effectiveness of a ground and air resource suppression organization in fighting forest wildfires. During a simulation, WIRAS needs to evaluate the opportunities for dispatching up to 100 suppression resources to numerous fire locations. This need for evaluation can happen many times during a single simulation leading to tens of thousands of travel time estimates. To be effective, the WIRAS simulation model requires computations to occur within a fraction of a second while providing a limited loss of accuracy on estimates of travel time. To accomplish this objective, we first explored the possibility of computing travel times over a network using the Dijkstra shortest-path algorithm (Dijkstra 1959). While numerous enhancements have been made over the years to the Dijkstra algorithm (Zahn 1997), for our particular problem, tests using the Dijkstra algorithm as implemented within ArcInfo (ESRI 2001) proved much too slow. Therefore, we decided to develop COSTPATHSQ, a new variation on the implementation of the Dijkstra algorithm that could exploit the particular nature of our travel time problem and meet the computational requirements for the WIRAS simulation model. Methods COSTPATHSQ uses the Dijkstra algorithm to estimate the least-cost path, where cost refers to travel time in this instance. In our model, travel is performed on a raster cost surface or grid rather than a vector network because wildfires do not necessarily fall on network features, such as roads and trails. A network could be constructed across the whole landscape, but the file size would be excessive and we believe the processing time would be prohibitively slow. A raster cost surface can be viewed as a regular triangular network with a network node at the center of every cell in the raster and network links radiating from each node to each of the eight adjacent nodes, horizontally, vertically, and diagonally. The main enhancement to the Dijkstra algorithm described here is the construction and employment of a multi-resolution cost surface, or cost pyramid. The cost pyramid allows COSTPATHSQ to initially and quickly move across a coarse resolution surface. The algorithm employs finer resolution surfaces to more accurately locate the end points (fig. 1). Depending on computational tolerance and accuracy needs, the user of COSTPATHSQ determines the number successive levels of resolution desired. Figure 1—How least-cost paths are constructed using a cost pyramid. USDA Forest Service Gen. Tech. Rep. PSW-GTR-xxx. xxxx. Proceedings of the Second International Symposium on Fire Economics, Planning, and Policy: A Global View 593 Session Poster— A Fast Method for Calculating Emergency Response Times —Hatfield, Wiitala, Wilson, and Levy Cost Pyramid The COSTPATHSQ algorithm determines for a cost surface the least-cost path between points using the Dijkstra algorithm in a manner similar to its implementation in ArcInfo (ESRI 2001). As mentioned above, COSTPATHSQ enhances the Dijkstra algorithm by employing the use of a multi-resolution cost surface, or cost pyramid. The cost pyramid is constructed by successively aggregating blocks of four cells in an upward progression through the surfaces (fig. 1). The first or bottom level of the pyramid is constructed of the original cells in the cost surface. These cells are grouped into fours to form new cells on the second level of the pyramid. Within each second-level cell, the first-level cell that is deemed to be the best pathway is selected as the center or hub of the second-level cell. The cost to travel between adjacent second-level cells is calculated between the cells’ hubs using the Dijkstra algorithm to travel on the first level of the pyramid. These eight travel costs are stored in each of the cells in the second level of the pyramid. Subsequently the second-level cells are grouped into fours to form cells on the third level of the pyramid (the highest level in fig. 1). Within each third-level cell, the second-level cell that is deemed to be the best pathway is selected as the hub, and the cost between third-level cells is calculated using the Dijkstra algorithm to travel on the second level of the pyramid. This process is repeated until only two rows and/or columns of cells remain. The result is increasingly generalized cost surfaces that retain much of the accuracy of the source data. This system of cost surfaces allows COSTPATHSQ to estimate the leastcost path between points in a fraction of a second with a limited loss of accuracy. Pathways The determination of which cells make better pathways is done prior to constructing the pyramid and is based upon the underlying transportation network. It is accomplished by selecting the one cell from the entire raster that bears the lowest cost and traveling the least-cost path to this origin cell from every other cell in the cost surface. If more than one cell bears the lowest cost, then one of these cells is selected at random to be the origin cell. The result of this preprocessing is a pathway grid. The value of each cell in the pathway grid is the number of times that that cell is crossed when traveling from each cell in the grid to the origin cell. Cells that receive higher values are deemed to be better pathways. Thus when four first-level cells are grouped to form a second-level cell, the first-level cell with the highest value in the pathway grid is selected as the hub of the second-level cell. This maximum value is assigned to the second-level cell and is used during the determination of hubs on the third-level of the pyramid, and so on.

These empirical results offered further evidence that consumers’ perception and emotional evaluation of color plays such an important role in decision making that it has a noteworthy influence on consumer’s evaluation about the price of an automobile. Furthermore, the present study provided quantitative information regarding the relationship between color and price with automobile purchase. There are many vehicles in the market that the color is a very important decision making criterion, and consumers are willing to pay more for a car of choice simply because of its color. This implies that purely emotional attributes can be a very powerful tool for pricing strategies of automobile sales. It is interesting to explore how these differences due to color are influenced by consumer characteristics (e.g., gender or age), regions (e.g., urban versus rural), and how they interact with vehicle types (e.g., luxury versus non-luxury). Further investigation in these areas will help us better understand the relationship between the emotional attribute and purchase price decision. (CCCW’s) are a potentially serious threat to brands. At least 50% of Fortune 1,000 companies have been targeted by CCCW’s. Extant studies provide useful descriptions, but surprisingly, related empirical research does not exist. This paper attempts to fill this gap by empirically investigating effects on consumers’ behavioral intentions, and the potential protection of strong branding. Results indicate a significant interaction between exposure to CCCW and strong branding, suggesting that CCCW’s reduce consumer perceptions, but strong brands offer some protection. Consistent with reactance theory, if the targeted brand is among a consumer’s favorites, perceptions are ironically increased.

SUMMARYDespite the popularity of electronic reverse auctions, limited empirical research on their use and key outcomes such as supplier cooperation, purchase price reductions and time savings has been conducted. The current research examines the role that the relative strategic importance of the existing buyer–supplier relationship plays in reverse auction usage, including the type of governance structure established and the three previously mentioned outcomes. Results indicated that existing buyer/supplier relationships characterized as strategic opted for relational forms of governance in reverse auctions. When using transactional governance, this research provides empirical support for reports that reverse auctions result in reduced purchase prices and increased purchaser productivity. Managerial implications are offered, along with directions for future research.