Heuristic Artificial Intelligence Research Articles

Optimization of Alpha-Beta pruning based on heuristic algorithm

Games have become an important place for testing Artificial Intelligence (AI). Minimax and Alpha-Beta Pruning are two common and basic algorithms implemented in game AIs. However, there are still limitations of the searching time and searching depth. This paper strives to improve the game AI with a Heuristic Algorithm to optimize both the searching time and depth. The experiment consists of three AIs built with Minimax, Alpha-Beta Pruning, and Heuristic Algorithm to evident the improvement. These AIs are built to play a traditional Chinese zero-sum game, Gobang, which can be seen as an enhanced version of tic-tac-toe but more advanced. The data is collected when AIs compete with each other. Comparing the search time of three AIs, there is significant improvement of AI implemented Heuristic Algorithm; the median search time for Heuristic AI is only half of the Alpha-Beta Pruning AI, and only quarter of the Minimax AI. Moreover, because the search time decreases, the searching depth of the Heuristic AI can also be increased. With the larger searching depth, the Heuristic AI also gains a higher winning rate against the other two AIs.

Biosimulation Using the Cellworks Computational Omics Biology Model (CBM) Identifies Novel Biomarkers to Inform Mitoxantrone, Etoposide, and Cytarabine (MEC)-Based Combination Therapy in Refractory & Relapsed Acute Myeloid Leukemia (AML) Patients

Background: The optimal treatment strategy for managing Acute Myeloid Leukemia (AML) and the use of reliable and predictive biomarkers to guide selection of cytotoxic chemotherapy regimens among patients with diverse genomic profiles remain unmet needs in the clinic. The combination of MEC [mitoxantrone (MIT), etoposide (VP16), and cytarabine (ARA-C)] is a commonly used regimen for relapsed or refractory AML patients. Unfortunately, many patients do not respond to MEC, and which of the three drug agents matters most for each individual patient is not known. Predictors of response are needed urgently.Methods: The Computational Omics Biology Model (CBM) is a computational multi-omic biology software model created using artificial intelligence heuristics and literature sourced from PubMed to generate a patient-specific protein network map. The aberration and copy number variations from individual cases served as input into the CBM. Disease-biomarkers unique to each patient were identified within patient-specific protein network maps. Biosimulations were conducted on the Cellworks Biosimulation Platform by measuring the effect of chemotherapy on a cell growth score comprised of a composite of cell proliferation, viability, apoptosis, metastasis, and other cancer hallmarks. Biosimulation of drugs was conducted by mapping the interaction of various drug combinations with the patient's genomic and pathway alterations based on signaling pathway mechanisms and their phenotypic consequences.The Cellworks Biosimulation Platform identified unique chromosomal signatures that permit a stratification of patients that are most likely to respond to MIT, VP16, or ARA-C as well as their combinations.65 AML patients were selected for this study largely based on genomic data published in TCGA and PubMed: •ARA-C [N=12, 7 responders (R) & 5 non-responders (NR)]•ARA-C + MIT [N=30, 29 R & 1 NR]•ARA-C + MIT + VP16 [N=23, 12 R & 11 NR]Results: Of the12 patients treated with ARA-C alone, 5 were predicted to be NR and 7 were predicted to be R. Of the 5 NR, 4 had 5q del which resulted in loss of APC, CSNK1A1 and SLC22A4 (nucleoside carrier) forming the non-response biomarkers for ARA-C. Notably, the biosimulation predicted lenalidomide to be beneficial for these patients. Out of 7 R, 4 patients also had 5q del, but were predicted to be R because of co-occurring aberrations involving CLSPN del, DHODH del, MSH2 del, EP300 del, CREBBP del, MSH6 Del, and RRM2 del. These genes were exclusively present in ARA-C responders. Of 53 patients who received ARA-C + VP16 + MIT or ARA-C + MIT, 41 patients were predicted to be R and 12 patients were predicted to be NR. The genomic aberrations predicted by biosimulation to be associated with response to this regimen include: NPM1-mut, TET2-mut, IDH1-mut, IDH2-mut, RAD17-del, NRAS-mut (Table 1). Notably, CBM predicted 19 of the 41 R had no genomic biomarkers of response to VP16 or MIT, suggesting these patients might have benefited equally from ARA-C alone with less toxicity and cost. Finally, 11/65 patients were predicted NR to MEC treatment. In the biosimulation, treatment failure was associated with high aberration frequencies of KMT2C-mut/del, FLT3 mut, TWIST1 del, LIMK1 del, SNAI2 amp, FNTA amp, and KAT6A amp. Of note, these genomic markers suggested a likelihood of benefit from other therapies, including vincristine, JQ1 and rigosertib.Conclusions: The Cellworks Biosimulation Platform identified novel polygenic biomarkers of response that can be employed to determine the optimal therapy for relapsed AML patients. Biosimulation permits avoidance of cytotoxic drugs with little chance of efficacy and reveals vulnerabilities in each patient's cancer that can be exploited to improve disease control. In AML, biosimulation promises to improve intensive therapy regimens by tailoring chemotherapy to optimize disease control and minimize toxicity. [Display omitted] DisclosuresMarcucci: Novartis: Other: Speaker and advisory scientific board meetings; Agios: Other: Speaker and advisory scientific board meetings; Abbvie: Other: Speaker and advisory scientific board meetings. Kumar: Cellworks Group Inc.: Current Employment. Castro: Cellworks Group Inc.: Current Employment; Omicure Inc: Consultancy; Caris Life Sciences Inc.: Consultancy; Exact sciences Inc.: Consultancy; Bugworks: Consultancy; Guardant Health Inc.: Speakers Bureau. Grover: Cellworks Group Inc.: Current Employment. Patil: Cellworks Group Inc.: Current Employment. Alam: Cellworks Group Inc.: Current Employment. Azam: Cellworks Group Inc.: Current Employment. Mohapatra: Cellworks Group Inc.: Current Employment. Tyagi: Cellworks Group Inc.: Current Employment. Kumari: Cellworks Group Inc.: Current Employment. Prasad: Cellworks Group Inc.: Current Employment. Nair: Cellworks Group Inc.: Current Employment. Lunkad: Cellworks Group Inc.: Current Employment. Joseph: Cellworks Group Inc.: Current Employment. G: Cellworks Group Inc.: Current Employment. Chauhan: Cellworks Group Inc.: Current Employment. Basu: Cellworks Group Inc.: Current Employment. Behura: Cellworks Group Inc.: Current Employment. Ghosh: Cellworks Group Inc.: Current Employment. Husain: Cellworks Group Inc.: Current Employment. Mandal: Cellworks Group Inc.: Current Employment. Raman: Cellworks Group Inc.: Current Employment. Patel: Cellworks Group Inc.: Current Employment. Mundkur: Cellworks Group Inc: Current Employment. Christie: Cellworks Group Inc.: Current Employment. Macpherson: Cellworks Group Inc.: Current Employment. Howard: Servier: Consultancy; Cellworks Group Inc.: Consultancy; Sanofi: Consultancy, Other: Speaker fees.

Biosimulation Using the Cellworks Computational Omics Biology Model (CBM)-Based Assessment of Cytarabine (ARA-C) and Anthracycline Response and Novel Biomarker Response Criteria for the Addition of Etoposide (VP16) in Acute Myeloid Leukemia (AML)

Biosimulation Using the Cellworks Computational Omics Biology Model (CBM)-Based Assessment of Cytarabine (ARA-C) and Anthracycline Response and Novel Biomarker Response Criteria for the Addition of Etoposide (VP16) in Acute Myeloid Leukemia (AML)

Therapy Biosimulation Using the Cellworks Computational Omics Biology Model (CBM) Is Predictive of Individual Acute Myeloid Leukemia (AML) Patient Probability of Clinical Response (CR) and Overall Survival (OS): Mycare-023

Therapy Biosimulation Using the Cellworks Computational Omics Biology Model (CBM) Is Predictive of Individual Acute Myeloid Leukemia (AML) Patient Probability of Clinical Response (CR) and Overall Survival (OS): Mycare-023

Biosimulation Using the Cellworks Computational Omics Biology Model (CBM) Identifies Immune Modulation As a Key Pathway for Predicting Azacitidine (AZA) Response in Myelodysplastic Syndromes (MDS)

Background: DNA methyltransferase inhibition (DNMTi) with the hypomethylating agents (HMA) azacitidine (AZA) or decitabine, remains the mainstay of therapy for the majority of high-risk Myelodysplastic Syndromes (MDS) patients. Nevertheless, only 40-50% of MDS patients achieve clinical improvement with DNMTi. There is a need for a predictive clinical approach that can stratify MDS patients according to their chance of benefit from current therapies and that can identify and predict responses to new treatment options. Ideally, patients predicted to be non-responders (NR) could be offered alternative strategies while being spared protracted treatment with HMA alone that has a low likelihood of efficacy. Recently, an intriguing discovery of immune modulation by HMA has emerged. In addition to the benefits of unsilencing differentiation genes and tumor suppressor genes, HMA's reactivate human endogenous retroviral (HERV) genes leading to viral mimicry and upregulation of the immune response as a major mechanism of HMA efficacy. Although the PD-L1/PD1 blockade plus HMA has been recognized as a beneficial combination, there are no established markers to guide decision-making. We report here the utility of immunomic profiling of chromosome 9 copy number status as a significant mechanism of immune evasion and HMA resistance.Methods: 119 patients with known clinical responses to AZA were selected for this study. Publicly available data largely from TCGA and PubMed was utilized for this study. The aberration and copy number variations from individual cases served as input into the Cellworks Computation Omics Biology Model (CBM), a computational biology multi-omic software model, created using artificial intelligence heuristics and literature sourced from PubMed, to generate a patient-specific protein network map. Disease-biomarkers unique to each patient were identified within protein network maps. The Cellworks Biosimulation Platform has the capacity to biosimulate disease phenotypic behavior and was used to create a disease model and then conduct biosimulations to measure the effect of AZA on a cell growth score comprised of a composite of cell proliferation, viability, apoptosis, metastasis, and other cancer hallmarks. Biosimulation of drug response was conducted to identify and predict therapeutic efficacy.Results: Although AZA treatment increased tumor associated antigens and interferon signaling, it also increased PD-L1 expression to inactivate cytotoxic CD8(+) T cells. Copy number alterations of the chromosome 9p region were found to significantly drive PD-L1 expression with multiple genes such as CD274, IFNA1, IFNA2, JAK2, PDCD1LG and KDM4C playing a role in PD-L1 regulation further increasing immune suppression (Figure 1). Among 6 cases of chromosome 9p aberration in this dataset, 9p amp (n=2) were clinical non-responders (NR) while 9p del (n=4) were responders (R) to AZA. In principle, checkpoint immunotherapy could improve outcomes for patients with 9p abnormalities. Additionally, copy number variation loss of key genes located on chromosome 16 involved in antigen processing and presentation such as CIITA, CTCF, IRF8, PSMB10, NLRC5, and SOCS1 were found to negatively impact AZA sensitivity (NR=4; R=0); these patients would also be unlikely to respond to checkpoint immunotherapy. Also, aberrations in melanoma antigen gene (MAGE) family proteins (NR=2; R=O), and STT3A (NR=1; R=5) were found to impact AZA efficacy by decreasing antigen processing on tumor cells.Conclusion: Based on the results from the Cellworks Biosimulation Platform applied to the CBM, copy number variants of chromosome 9p and 16 can be converted into CBM-derived biomarkers for response to checkpoint immunotherapy in combination with HMA. Our results support a future prospective evaluation in larger cohorts of MDS patients. [Display omitted] DisclosuresHoward: Servier: Consultancy; Cellworks Group Inc.: Consultancy; Sanofi: Consultancy, Other: Speaker fees. Kumar: Cellworks Group Inc.: Current Employment. Castro: Bugworks: Consultancy; Exact sciences Inc.: Consultancy; Guardant Health Inc.: Speakers Bureau; Cellworks Group Inc.: Current Employment; Caris Life Sciences Inc.: Consultancy; Omicure Inc: Consultancy. Grover: Cellworks Group Inc.: Current Employment. Mohapatra: Cellworks Group Inc.: Current Employment. Kapoor: Cellworks Group Inc.: Current Employment. Tyagi: Cellworks Group Inc.: Current Employment. Nair: Cellworks Group Inc.: Current Employment. Suseela: Cellworks Group Inc.: Current Employment. Pampana: Cellworks Group Inc.: Current Employment. Lala: Cellworks Group Inc.: Current Employment. Singh: Cellworks Group Inc.: Current Employment. Shyamasundar: Cellworks Group Inc.: Current Employment. Kulkarni: Cellworks Group Inc.: Current Employment. Narvekar: Cellworks Group Inc.: Current Employment. Sahni: Cellworks Group Inc.: Current Employment. Raman: Cellworks Group Inc.: Current Employment. Balakrishnan: Cellworks Group Inc.: Current Employment. Patil: Cellworks Group Inc.: Current Employment. Palaniyeppa: Cellworks Group Inc.: Current Employment. Balla: Cellworks Group Inc.: Current Employment. Patel: Cellworks Group Inc.: Current Employment. Mundkur: Cellworks Group Inc: Current Employment. Christie: Cellworks Group Inc.: Current Employment. Macpherson: Cellworks Group Inc.: Current Employment. Marcucci: Abbvie: Other: Speaker and advisory scientific board meetings; Novartis: Other: Speaker and advisory scientific board meetings; Agios: Other: Speaker and advisory scientific board meetings.

Biosimulation Using the Cellworks Computational Omics Biology Model (CBM) Predicted Novel Biomarkers for Hyper-CVAD (CVAD) Treatment Response and Combination of Rituximab and Cladribine (RC) in CVAD Resistant Cases of Mantle Cell Lymphoma (MCL)

Biosimulation Using the Cellworks Computational Omics Biology Model (CBM) Predicted Novel Biomarkers for Hyper-CVAD (CVAD) Treatment Response and Combination of Rituximab and Cladribine (RC) in CVAD Resistant Cases of Mantle Cell Lymphoma (MCL)

Estimation of total dissolved solids, electrical conductivity, salinity and groundwater levels using novel learning machines

In this study, the groundwater parameters including electrical conductivity (EC), salinity, total dissolved solids (TDS) and groundwater level (GWL) for a 15-year time series (from 2002 to 2017) of the Mighan Plain, located in Markazi Province, Iran, were simulated using a hybrid meta heuristic artificial intelligence (AI) model called “wavelet self-adaptive extreme learning machine” (WSAELM) for the first time. Initially, the detection of the most significant lags of the time-series data was conducted using the autocorrelation function (ACF) and the partial autocorrelation function (PACF) analyses. By using these lags, four WSAELM models were defined and then the superior models in simulating the TDS, EC, salinity and GWL were introduced. The values of the determination coefficient (R2), Variance Accounted for (VAF) and the Nash–Sutcliffe efficiency coefficient (NSC) for the superior model simulating salinity were computed to be 0.991, 98.124 and 0.980, respectively. Also, approximately 44% of the TDS values modeled by the best model had an error less than 10%, while roughly a third of the TDS values estimated by the model had an error more than 15%. The findings indicated that the proposed hybrid model underestimated the GWL parameter, while it performed in an overestimate way for other parameters. The results of the uncertainty analysis showed the low width of uncertainty for GWL (WUB = ± 0.798), TDS (WUB = ± 5.035), EC (WUB = ± 6.425) and salinity (WUB = ± 66.650).

Recent Trends and Development of Heuristic Artificial Intelligence Approach in Mechanical System and Engineering Product Design

Recent Trends and Development of Heuristic Artificial Intelligence Approach in Mechanical System and Engineering Product Design

Modeling of Seepage Flow Through Concrete Face Rockfill and Embankment Dams Using Three Heuristic Artificial Intelligence Approaches: a Comparative Study

The goal was to predict seepage flow (Q) through concrete face rockfill and embankment dams, using three artificial intelligence models, i.e., multivariate adaptive regression splines (MARS), least squares support vector machine (LSSVM), and M5 model tree (M5Tree). The three models were constructed exclusively using in situ measured data from two dams: El Agrem dam located at Jijel province, and Fontaine Gazelles dam located at Biskra province. The obtained results using artificial intelligence models were compared to those obtained using the multiple linear regression (MLR) models. We used two different input variables for developing the models: (i) the daily reservoir water level (WL) and the piezometer elevation (PL) measured at seven different piezometers (PZ1 to PZ7). The results show that the estimation accuracy for Fontaine Gazelles dam is much better than those obtained for El Agrem. All the models performed reasonably well, but the LSSVM was the most consistent predictor of seepage flow for the two data sets. The validation results showed that the LSSVM model has showed significantly better accuracy of seepage flow prediction with root mean square error (RMSE) of 0.432 L/s, mean absolute error (MAE) of 0.302 L/s and correlation coefficient R of 0.952 for Fontaine Gazelles, and RMSE of 0.544 L/s, MAE of 0.344 L/s and correlation coefficient R of 0.731 for El Agrem dam. From this study we conclude that, seepage flow is likely to vary considerably, depending on the reservoir water level, and that the proposed model can be very helpful in estimation of seepage flow, while limitations of the prediction using a standard regression model are illustrated.

Fuzzy modeling of dependability optimization for supporting the production-quality strategies - case study in technical field

this work treats dependability from a functional (systemic) standpoint, which implies meeting a function required within an input/output system. This approach is demanded and necessary within technical equipment design stages, and is an integrant part of systems integrity design methodology. Designing the system integrity includes design criteria for reliability, availability, maintainability and safety of any system and equipment. The combination of these four concepts leads to the necessity of a comparative and integrative methodology that should ensure a good systems design, with required integrity values that can be computed easily, analyzed the most complete possible, and with the possibility to be modified accordingly. During the recent years, artificial intelligence techniques have been developed for dependability, that make use of, statistic methods (necessary for the realization of an operation history), deterministic mathematic algorithms (to determine the exact solutions when these can be determined and are required and the time resource is not critical), heuristic artificial intelligence methods (that can provide good quality solutions when time is a critical resource) - neuronal networks, genetic algorithms (to determine predicted or extreme values for system variables), as well as fuzzy methods for systems in which the system variables change within certain values intervals. All these techniques offer a global image of the artificial intelligence modelling (AIM) in designing the reliability, availability, maintainability and safety, in order to offer a continuous design feedback mean during the entire engineering design process. In this work, we shall develop a fuzzy model for a technical process and will compare our results with the results supplied by mathematical, neuronal and genetic models applied to the same system. The qualitative comparison of models will be completed with their quantitative comparison, by analyzing the complexity of algorithms and execution time. The utility of this work consists in the implementation of an inter-disciplinary tool easy to apply to technical systems with measurable dependability variables that can provide during the flow sheet, the values requested from technical and economic standpoint.

Efficient Online Portfolio Selection with Heuristic AI Algorithm

Online portfolio selection is one of the most important problems in several research communities, including finance, engineering, statistics, artificial intelligence, and machine learning, etc. The primary aim of online portfolio selection is to determine portfolio weights in every investment period (i.e., daily, weekly, monthly etc.) to maximize the investor’s final wealth after the end of investment period (e.g., 1 year or longer). In this paper, we present an efficient online portfolio selection strategy that employs a heuristic artificial intelligence (AI) algorithm to maximise the total wealth based on historical stock prices. Based on empirical studies conducted on recent historical datasets for the period 2000 to 2017 on four different stock markets (i.e., NYSE, S&P500, DJIA, and TSX), the proposed algorithm has been shown to outperform both Anticor and OLMAR —the two most prominent portfolio selection strategies in contemporary literature. The algorithm achieved 34.22 of the total wealth while Anticor and OLMAR returned 1.08, and 4.52 respectively, on the NYSE market. On the S&P market, the algorithm returned 15.39 of the total wealth while the Anticor and OLMAR returned 6.20 and 2.2,respectively. On the TSX market, the algorithm returned 2.43 of the total wealth while Anticor and Olmar returned 0.96 and 0.41, respectively.

Long‐term modelling of wind speeds using six different heuristic artificial intelligence approaches

Wind speed is an essential component that needs to be determined accurately, especially over long‐term periods for various engineering and scientific purposes including renewable energy productions, structural building sustainability and others. In this study, six different heuristic methods: multi‐layer perceptron artificial neural networks, (ANN), adaptive neuro‐fuzzy inference system (ANFIS) with grid partition (GP), ANFIS with subtractive clustering (SC), generalized regression neural networks (GRNN), gene expression programming (GEP) and multivariate adaptive regression spline (MARS) are developed to model monthly wind speeds using meteorological input information. The atmospheric pressure, temperature, relative humidity and rainfall values are obtained from Jolfa and Tabriz meteorological stations, Iran, and are used to build the proposed predictive models.. Different statistical indicators are computed to evaluate and comprehensively assess the performance of the six heuristic methods. Over the testing phase, the ANFIS‐GP and GRNN models are seen to exhibit the highest predictive performance for the Jolfa and Tabriz stations, respectively. That is, the maximum coefficient of determination are found to be 0.874, 0.858, 0.850, 0.849, 0.847 and 0.826, for the GRNN, ANFIS‐GP, ANFIS‐SC, ANN, GEP and MARS models, respectively, for Jolfa station, respectively, revealing the superiority of GRNN over the five counterpart models. The results show the generalization capability of the tested heuristic artificial intelligence techniques for both study stations, and therefore could be explored for windspeed prediction and various decisions made in regards to climate change studies.

Backtracking Algorithm-based Disassembly Sequence Planning

Disassembly sequence planning (DSP) is one of the core research themes of product life cycle design. Effective DSP can reduce product life cycle costs as well as their environmental effects and provides a theoretical future foundation for recycling and improving sustainability performance. The problem with current heuristic artificial intelligence algorithms is that the outcome may have a local extremum; therefore, it may not generate the exact or even the approximate optimum solution. For a specific sequence planning problem, some algorithm parameters need to be adjusted in order to generate a perfect solution. In this study, the objective is to create new DSP by integrating a constraint satisfaction problem (CSP) based on backtracking algorithms. First, the relationship between DSP and the CSP was analysed. Then, the DSP was converted into a CSP based on a disassembly constraint graph (DCG). The solution flowchart of DSP and the CSP-associated algorithms, based on backtracking algorithms, are presented in this article for a dishwasher recycling case study. The outcome of this study shows that the created method can effectively find all of the feasible disassembly sequences as well as the optimum sequence for recycling. Ultimately, these results can be effectively applied to more appliances for future use.

Optimal Capacitor Placement Techniques in Transmission and Distribution Networks to Reduce Line Losses and Voltage Stability Enhancement: A Review

The various optimal capacitor placement techniques on transmission and distributions lines for line losses reduction and enhancement of voltage stability in the power system network have been proposed so far in different papers. Optimal Capacitor placement is an optimization problem which has an objective to define the sizes and locations of capacitors to be installed. This paper focuses on the optimal capacitor placement and sizing problem formulation and analytical as well as heuristic artificial intelligence optimization methods for optimal capacitor placement and sizing. This paper helps the researchers to know about the different methods presented so far for optimal capacitor placement and sizing in transmission and distribution networks so that further work on optimal capacitor placement can be carried out for better results. Keywords - capacitor sizing, distribution system, heuristic techniques, optimal capacitor placement, reactive power, transmission system, voltage stability.

Evaluating artificial intelligence heuristics for a flexible Kanban system: simultaneous Kanban controlling and scheduling

This paper evaluates several artificial intelligence heuristics for a simultaneous Kanban controlling and scheduling on a flexible Kanban system. The objective of the problem is to minimise a total production cost that includes due date penalty, inventory, and machining costs. We show that the simultaneous Kanban controlling and scheduling is critical in minimising the total production cost (approximately 30% cost reduction over scheduling without a Kanban controlling). To identify the most effective search method for the simultaneous Kanban controlling and scheduling, we evaluated widely known artificial intelligence heuristics: genetic algorithm, simulated annealing, tabu search, and neighbourhood search. Computational results show that the tabu search performs the best in terms of solution quality. The tabu search also requires a much less computational time than the genetic algorithm and the simulated annealing. To further improve the solution quality and computational time for a simultaneous Kanban controlling and scheduling on a flexible Kanban system, we developed a two-stage tabu search. At the beginning of the tabu search process, an initial solution is constructed by utilising the customer due date information given by a decision support system. The two-stage tabu search performs better than the tabu search with a randomly generated initial solution in both the solution quality and computational time across all problem sizes. The difference in the solution quality is more pronounced at the early stages of the search.

Artificial intelligence heuristics in solving vehicle routing problems with time window constraints

This paper describes the authors’ research on various heuristics in solving vehicle routing problem with time window constraints (VRPTW) to near optimal solutions. VRPTW is NP-hard problem and best solved to near optimum by heuristics. In the vehicle routing problem, a set of geographically dispersed customers with known demands and predefined time windows are to be served by a fleet of vehicles with limited capacity. The optimized routines for each vehicle are scheduled as to achieve the minimal total cost without violating the capacity and time window constraints. In this paper, we explore different hybridizations of artificial intelligence based techniques including simulated annealing, tabu search and genetic algorithm for better performance in VRPTW. All the implemented hybrid heuristics are applied to solve the Solomon's 56 VRPTW with 100-customer instances, and yield 23 solutions competitive to the best solutions published in literature according to the authors’ best knowledge.

Linear gate assignment: a fast statistical mechanics approach

This paper deals with the problem of linear gate assignment in two layout styles: one-dimensional logic array and gate matrix layout. The goal is to find the optimal sequencing of gates in order to minimize the required number of tracks, and thus to reduce the overall circuit layout area. This is known to be an NP-hard optimization problem, for which no absolute approximation algorithm exists. Here we report the use of a new optimization heuristic derived from statistical mechanics-the microcanonical optimization algorithm, /spl mu/0-to solve the linear gate assignment problem. Our numerical results show that /spl mu/0 compares favorably with at least five previously employed heuristics: simulated annealing, the unidirectional and the bidirectional Hong construction methods, and the artificial intelligence heuristics GM Plan and GM Learn. We also show how the algorithm is able to outperform microcanonical annealing. Moreover, in a massive set of experiments with circuits whose optimal layout is not known, our algorithm has been able to match and even to improve, by as much as seven tracks, the best solutions known so far.

Dynamic Risk Management Systems: Hybrid Architecture and Offshore Platform Illustration

This paper describes and illustrates the architecture of computer‐based Dynamic Risk Management Systems (DRMS) designed to assist real‐time risk management decisions for complex physical systems, for example, engineered systems such as offshore platforms or medical systems such as patient treatment in Intensive Care Units. A key characteristic of the DRMSs that we describe is that they are hybrid, combining the powers of Probabilistic Risk Analysis methods and heuristic Artificial Intelligence techniques. A control module determines whether the situation corresponds to a specific rule or regulation, and is clear enough or urgent enough for an expert system to make an immediate recommendation without further analysis of the risks involved. Alternatively, if time permits and if the uncertainties justify it, a risk and decision analysis module formulates and evaluates options, including that of gathering further information. This feature is particularly critical since, most of the time, the physical system is only partially observable, i.e., the signals observed may not permit unambiguous characterization of its state. The DRMS structure is also dynamic in that, for a given time window (e.g., 1 day or 1 hour), it anticipates the physical system's state (and, when appropriate, performs a risk analysis) accounting for its evolution, its mode of operations, the predicted external loads and problems, and the possible changes in the set of available options. Therefore, we specifically address the issue of dynamic information gathering for decision‐making purposes. The concepts are illustrated focusing on the risk and decision analysis modules for a particular case of real‐time risk management on board offshore oil platforms, namely of two types of gas compressor leaks, one progressive and one catastrophic. We describe briefly the DRMS proof‐of‐concept produced at Stanford, and the prototype (ARMS) that is being constructed by Bureau Veritas (Paris) based on these concepts.

Computer-Controlled Manipulator With Visual Inputs

A computer control algorithm utilizing the modern control theory and the heuristic artificial intelligence approach has been developed to control the MIT Scheinman electric arm. The control algorithm is task-oriented with the capabilities of accepting visual input coordination and discrete word voice commands in addition to the linguistic commands typed in by an operator at the remote terminal. The motion of the arm is considered to be composed of motion of the wrist and orientation of the hand. A human operator, always present in the control loop and interacting with the system, generates linguistic task-directed commands at the remote terminal. A task-directed command is recognized, interpreted, and decoded into sequence of subtasks. The first subtask corresponds to the motion of the wrist which is controlled by the suboptimal feedback controller. The remaining four subtasks are further broken down into combinations of six primitive movements which govern the position/orientation of the hand. The objective of the visual recognition algorithm is to identify ob-jects and their locations surrounding the arm from its environmental library or model. The library is then updated to initiate the arm to complete its execution of task. Areas and circumferences of the objects are the two features chosen for recognition. The recognition of an object is then based on the threshold value of the weighted sum of area and circumference of the object. The set of weights is trained again when a new object is introduced to the arm and incorporated into the library. The real-time implementation of the algorithm on AARL MIT arm connected to a PDP 11/45 computer shows that it can recognize the objects surrounding the arm within 50 seconds.