Scheduling Of Agents Research Articles

Epoetin Alfa and Darbepoetin Alfa Go Head to Head

Epoetin Alfa and Darbepoetin Alfa Go Head to Head

Modeling Uncertainty and its Implications to Sophisticated Control in Tæms Agents

Open environments are characterized by their uncertainty and non-determinism. Agents need to adapt their task processing to available resources, deadlines, the goal criteria specified by the clients as well their current problem solving context in order to survive in these environments. If there were no resource constraints, then an optimal Markov Decision Process based policy would obviously be the best way for complex problem solving agents to make scheduling decisions. However in many agent systems, these scheduling decisions have to be made on-line or in soft real-time, making the off-line policy computationally infeasible in open environments. The hybrid planner/scheduler used to control Task Analysis, Environment Modeling, and Simulation (TAEMS) agents is the Design-to-Criteria (DTC) agent scheduler. Design-to-Criteria scheduling is the soft real-time process of custom building a plan/schedule to meet an agent's current objectives which are expressed as dynamic goal criteria (including real-time deadlines), using task models that describe alternate ways to achieve tasks and subtasks. Recent advances in Design-to-Criteria control include the addition of uncertainty to the TAEMS computational task models analyzed by the scheduler and the incorporation of uncertainty in the scheduling process. As we show, the use of uncertainty in TAEMS and Design-to-Criteria enables agents to make better control decisions in uncertain environments. Design-to-Criteria uses a heuristic approach for on-line scheduling of medium granularity tasks.It approximates the analysis used to generate an optimal policy by heuristically reasoning about the implications of uncertainty in task execution. The addition of uncertainty has also spawned a post-scheduling contingency analysis step for situations in which an agent must produce a result by a given deadline (deadline critical situations) and where the added computational cost is worth the expense. We describe the uncertainty representation in TAEMS and how it improves task models and the scheduling process, and provide empirical examples of reasoning about uncertainty in action. We also evaluate the performance of our heuristic-based approach to agent control using the performance of the policy generated by an optimal controller as the benchmark.

Towards a generic distributed and collaborative digital manufacturing

A framework for distributed manufacturing is proposed to facilitate collaborative product development and production among geographically distributed functional agents using digitalized information. Considering the complexity of products created in a distributed manufacturing scenario, it often requires close collaborations among a number of facilities. In this research work, various functional agents, such as the manufacturability evaluation agent (MEA), manufacturing resource agent (MRA), process-planning agent (PPA), manufacturing scheduling agent (MSA), shop floor agent (SFA), fault diagnosis agent (FDA), etc., can interact coherently for distributed manufacturing. With specific agents having unique functionalities, a manufacturing managing agent (MMA) acts as the centre of this distributed manufacturing system. The MMA agent assists the specific agents’ to work seamlessly and also to collaborate closely with the participating agents. In this way, the production cycle of a part can be optimized from product design to final manufacturing since all the production procedures are considered logically and every procedure is correlated. The agent language based on the knowledge query manipulation language (KQML) includes many pre-defined performatives that ease the participating agents to carry out their tasks intelligently by interpreting commands from one another. Additionally, to ensure the adaptiveness and upgradeability of the system, the internal structure of each functional agent that is based on JATLite is modularized into several components, including a communication interface, central work engine, knowledge base pool, and input/output modifier for possible future methodology enhancements.

Distributed project scheduling with information sharing in supply chains: part II—theoretical analysis and computational study

The first part of this paper (Lau, J.S.K., Huang, G.Q. and Mak, K.L., Distributed project scheduling with information sharing in supply chains: part I—agent-based negotiation algorithm. Int. J. Prod. Res., 2005, 43, 4813–4838) has outlined the background of this research. The paper proposes and demonstrates a new negotiation-based algorithm (NEG) for solving distributed project scheduling problems (DPSP). This new algorithm not only acknowledges and accommodates the autonomy and independence of individual enterprises in making decisions in the entire supply chain, but also takes advantage of limited information shared among them to improve the quality and efficiency. This second paper will conduct a thorough theoretical analysis and computational study on the proposed algorithm. The theoretical analysis shows that the algorithm converges such that schedules of project and contractor agents do not have conflicts. The computational study reveals that NEG performs better than a centralized heuristic and contract net protocols in terms of the solution quality (e.g. total operating cost). Computational efficiency of NEG is also reasonably comparable and competitive to the extent that good quality solutions can be obtained within an affordable time limit.

Effect of the Selective Estrogen Receptor Modulator Arzoxifene on Repopulation of Hormone-Responsive Breast Cancer Xenografts between Courses of Chemotherapy

Selective inhibition of repopulation of clonogenic tumor cells between courses of chemotherapy has potential to improve the effectiveness of treatment. Here we study arzoxifene, a short-acting selective estrogen receptor modulator, for its potential to inhibit repopulation in estrogen-dependent human breast cancer MCF-7 xenografts between courses of chemotherapy. Proliferation of tumor cells was evaluated by cyclin D1 expression and uptake of 5-bromo-2'-deoxyuridine. Arzoxifene decreased cell proliferation in xenografts. To model adjuvant treatment of human breast cancer, MCF-7 cells were injected s.c. into nude mice and four groups of mice received the following treatments beginning after implantation: (a) control (vehicle solution); (b) arzoxifene alone, 5 days per week by oral gavage for 3 weeks; (c) 5-fluorouracil (5-FU) or paclitaxel i.p. weekly, for 3 doses; and (d) arzoxifene following each cycle of chemotherapy. The incidence of tumors with volume > or =50 mm(3) was determined as a function of time. MCF-7 xenografts developed in 100% of control mice by 4 weeks after implantation. Paclitaxel or 5-FU alone had minor effects to delay the appearance of xenografts whereas arzoxifene alone caused longer delay. Combined treatment with arzoxifene given between cycles of 5-FU or paclitaxel had substantial effects, with approximately 50% tumor incidence by 5 weeks. Our results indicate that arzoxifene can inhibit repopulation of hormone-responsive MCF-7 breast cancer xenografts when given between courses of chemotherapy. The scheduling of short-acting hormonal agents between courses of adjuvant chemotherapy for human breast cancer has potential to improve the outcome of treatment.

반도체 웨이퍼 팹의 에이전트 기반 스케쥴링 방법

This paper proposes an agent-based scheduling method for semiconductor wafer fabrication facilities with hard inter-operation temporal constraints. The scheduling problem is to find the feasible schedules that guarantee both logical and temporal correctness. A proposed multi-agent based architecture is composed of scheduling agents, workcell agents, and machine agents. A scheduling agent computes optimal schedules through bidding mechanisms with a subset or entire set of the workcell agents. A dynamic planning-based approach is adopted for the scheduling mechanism so that the dynamic behaviors such as aperiodic job arrivals and reconfiguration can be taken into consideration.

Just When You Thought the Fluorouracil Debate Was Over: S-1 and Gastric Cancer

Despite a decline in incidence in the United States and Western Europe, gastric cancer is common worldwide and remains the second leading cause of cancer-related death. In a shift in the epidemiology of gastric cancer, adenocarcinomas of the esophagus and gastroesophageal junction are increasing at a dramatic rate in the United States and Western Europe, and represent an even more virulent disease entity. The pattern of more distal gastric cancer elsewhere in the world may be a function of rates of infection by Helicobacter pylori, dietary practices, and potential dietary deficiency in micronutrients such as selenium and vitamin C. The predominance of adenocarcinoma of the esophagus and gastroesophageal junction in the West may reflect population trends in obesity and tobacco abuse, and paradoxically, a decrease in Helicobacter infection that may increase the risk of esophageal reflux disease. Because of the absence of effective screening, patients with gastric cancer in the United States usually present with symptomatic and either locally advanced or metastatic disease. High rates of recurrence after surgery have engendered decades of clinical trials evaluating the use of adjuvant chemotherapy, with or without radiation. Fortunately, recent clinical trials have made advances with the validation of the use of systemic chemotherapy to improve survival. Intergroup trial 116 showed that postoperative bolus fluorouracil (FU), leucovorin, and radiation therapy improved survival compared with surgery alone, despite the relative inactivity for FU and leucovorin in metastatic disease. Although postoperative adjuvant chemotherapy alone has failed to affect survival in gastric cancer, preand postoperative chemotherapy with epirubicin, cisplatin, and protracted-infusion FU (ECF) without the use of radiation, as recently reported in the MAGIC trial from the United Kingdom, significantly improved survival compared with surgery alone. The improvement in survival in recent adjuvant trials, however, is marginal, and the key to further progress is the identification of more active systemic agents in gastric cancer. The evaluation of a novel combination chemotherapy regimen, S-1 and cisplatin, is the focus of the phase I trial reported by Ajani et al in the current issue of the Journal of Clinical Oncology. This trial raises important issues about the pace of progress of chemotherapy development in gastric cancer, including the role of oral agents as a replacement for intravenous drugs. The trial also, for better or for worse, rekindles the time-worn debate of how best to administer fluoropyrimidines as part of combination chemotherapy. Conventional chemotherapy used in the treatment of gastric cancer has limited effectiveness and significant toxicity, depending on the dose and schedule of agents used. If we are searching for direction in gastric cancer drug development, colorectal cancer should clearly serve as a model. Progress has been made in colorectal cancer in the identification of new chemotherapy agents, optimal dosing and scheduling of agents, and the incorporation of targeted therapies. Colorectal cancer trials have firmly established that, in comparison with bolus FU, the use of continuous-infusion FU administered as a biweekly 48-hour infusion results in superior antitumor efficacy, longer patient survival, and improved therapy tolerance. The debate about the optimal schedule of FU has largely been replaced by discussion of the best sequencing of available therapies and the addition of targeted agents. The substitution of oral agents for continuous-infusion FU is a topic of more secondary interest given the accumulating data from phase III trials that show equivalence, but not necessarily superiority, of oral drugs. Whereas response, survival, and toxicity benefits have all been accomplished in colorectal cancer trials, such advances have eluded gastric cancer investigators. In gastric cancer, there has been a seeming reluctance to incorporate lessons learned from the trial experience in colorectal cancer. There is continued acceptance of the combination of a 5-day infusion of FU in combination with every 3to 4-week dosing of relatively high doses of cisplatin. The toxicity and limited efficacy of this regimen was underscored in a phase III JOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L VOLUME 23 NUMBER 28 OCTOBER 1 2005

Imatinib mesylate and its potential implications for gynecologic cancers

Among gynecologic malignancies, ovarian carcinoma is the most frequent cause of death, with the majority of patients presenting at advanced stage. There is a high rate of recurrence despite first-line chemotherapy. Sarcoma of the uterus, while accounting for a small percent of uterine cancers, is also associated with a high-recurrence rate and poor overall survival. Therefore, the development of novel treatment strategies is paramount. Imatinib mesylate (Gleevec; Novartis Pharmaceuticals Corp., East Hanover, NJ) is a tyrosine kinase inhibitor with activity against abl, c-kit, and platelet derived growth factor receptor (PDGFR), and is approved for the treatment of chronic myelogenous leukemia and gastrointestinal stromal tumor. Preclinical data provides evidence for c-kit and PDGFR expression in ovarian epithelial carcinomas and uterine sarcomas and have led to clinical trials evaluating the use of imatinib in these malignancies. Additionally, inhibition of PDGFR signaling has been proposed as an effective mechanism of chemotherapy by lowering tumor interstitial fluid pressure. Recent data have also suggested benefit with metronomic scheduling of cytotoxic agents at lower doses at more frequent dosing intervals, in combination with other targeted therapies. While activity of this agent remains to be established, further studies of imatinib in gynecologic malignancies are warranted, to demonstrate not only single-agent activity and the enhancement of cytotoxicity of other antineoplastic agents.

Collaborative agents for production planning and scheduling (CAPPS): a challenge to develop a new software system architecture for manufacturing management in Japan

Considering the frequent changes in market needs, real-time production management will produce huge additional value for manufacturers. As scheduling of shop floor operations is a local and partial decision in enterprises, the distributed local scheduling problems should be integrated at the enterprise level. Since decisions at the shop floor are very valuable, the next generation of production management will be performed as a decentralized system. This paper proposes architecture for scheduling-intensive production management. In the architecture, referred to as collaborative agents for production planning and scheduling, system integration is only achieved by communication among agents, i.e. a system is composed of loose connections between agents. As an implementation method, the paper introduces a standard specification for the communications among agents. The specification is recommended by PSLX, the Japan-based consortium for production planning and scheduling. The paper also shows some emerging business models on scheduling-intensive production management.

Internet Scheduling Environment With Market-Driven Agents

This paper describes a new generation scheduling paradigm, the Internet scheduling environment. It is formed by a group of Internet scheduling agents which share computational resources to solve scheduling problems in a distributed and collaborative manner. We propose a migration scheme to transform existing standalone scheduling systems to Internet scheduling agents that can communicate with each other and solve problems beyond individual capabilities. To coordinate computational resource collaboration among agents, we introduce the market-based control mechanism in which self-interested agents initiate or participate in auctions to sell or buy scheduling problems. Efficient allocation of computational resources is achieved through the auctions. This paper also describes a prototype Internet scheduling environment named LekiNET, which is migrated from LEKIN/spl reg/, a flexible job shop scheduling system. The experiments on the LekiNET testbed demonstrate that the agent-based market-driven Internet scheduling environment is feasible and advantageous to future scheduling research and development.

Constraints to feedback provision on forestry-related technologies

This paper ascertained the constraints to feedback provision on forestry-related technologies. Interview schedule was used to elicit information from 163 randomly selected respondents. Descriptive (frequencies, percentages) and inferential (Chi square and Ordinary Least square regression) statistics were used to analyse data collected for the study. Chi square statistics shows significant relationship between gender, age, marital status, number of wives, education and constraints to feedback provision. The major constraints that were identified as affecting feedback provision on forestry related technologies include illiteracy, unstable government policy, busy schedule of extension agents, busy schedule of researchers, low ratio of extension agents to farmers, poor infrastructure. The regression result showed R2 value of 0.67 which is an indication that 67 percent of constraints to feedback provision were explained by other independent variables and the multiple R of 0.82 shows a very strong correlation between the independent and dependent variables. Only farm size (t-value = -3.31, p

Feasibility of CT scan-guided Tru-Cut serial liver biopsies to evaluate pharmacodynamic endpoints in patients with liver metastasis treated with experimental drugs

We assessed the feasibility of using computed tomographically guided, Tru-Cut serial liver biopsies to perform pharmacodynamic studies in patients with liver metastasis who had been treated with experimental drugs. Twenty-three patients with liver metastasis were enrolled in two protocols for evaluation of the response to two new biologic drugs: a farnesyltransferase inhibitor and a tyrosine kinase inhibitor of the epidermal growth factor receptor. Computed tomographically guided Tru-Cut biopsies with an 18-gauge needle were performed. The procedure was performed at baseline and 2, 6, and 24 h after starting the administration of the drug. The procedures were scheduled on an outpatient basis. We obtained 271 samples (168 from metastatic lesions and 103 from surrounding normal liver tissue) from 23 patients. Biochemical determination of farnesyltransferase activity and different immunohistochemistry stainings (pEGRF, pMAPK, p27, Ki67, and apoptosis) of the signal transduction pathway were determined in each tissue sample, and all scheduled tissue assays were performed with the tissues obtained. Complete serial biopsies were performed in 20 patients. Three patients were excluded due to minor complications after the first biopsy. The morbidity rate in relation to the number of procedures was 1.1% (three of 271 samples). We found serial liver biopsies using Tru-Cut to be a safe and easy procedure. In our series, we could evaluate samples for molecular changes that influence the optimal dose, sequence, and schedule of the new antineoplastic agents.

Clinical studies in patients with solid tumors using a second-generation antisense oligonucleotide (GEM 231) targeted against protein kinase A type I.

GEM 231 is a second-generation antisense oligonucleotide targeted against the RIalpha regulatory subunit of cAMP-dependent protein kinase type I (PKA-I). Excessive expression of PKA-I is associated with cell proliferation and transformation, and increased levels of secreted extracellular PKA (ECPKA) are found in the serum of cancer patients. Preclinical studies have demonstrated single-agent antitumor activity of GEM 231 in a variety of human cancer xenograft models, and additive or synergistic antitumor activity has been observed with taxane and/or camptothecin-based combinations. Based on prior safety (MTD) data demonstrating dose-dependent, reversible, and cumulative transaminitis, and high peak plasma concentration (Cmax)-dependent changes in activated partial thromboplastin time (aPTT) with GEM 231 2-h twice-weekly infusions, an alternative schedule of GEM 231 given as a single agent was evaluated in patients with advanced solid tumors. Fourteen patients (median age approximately 60 yrs) with advanced solid malignancies received a total of 78 weeks of therapy. GEM 231 was infused via a CADD pump at 80 mg/m2/day (d) for 3 d/wk (n = 1), then for 5 d/wk at 80 (n = 3), 120 (n = 8), and 180 mg/m2/d (n = 2). One cycle was defined as 4 weeks of therapy. Apparent dose dependency for the occurrence of transaminitis was readily reversible. At 180 mg/m2/d, 2 of 2 patients had cycle 1 dose-limiting toxicity (DLT) transaminitis. One patient treated at 120 mg/m2/d experienced grade 3 transaminase elevations after 8 weeks of therapy, but when serum transaminase values rapidly improved he resumed treatment at 80 mg/m2/d for 6 weeks until tumor progression was documented. Another patient at 120 mg/m2/d developed grade 3 esophagitis after 3 weeks, limiting further dosing. One patient (lung cancer) demonstrated stable disease for 9 weeks. Overall, plasma aPTT was minimally prolonged and changes were transient, peaked at the end of each infusion, and were not associated with spontaneous bleeding. A constitutive symptom (e.g., low-grade fatigue) was common, cumulative, and reversible following discontinuation of therapy. Serum ECPKA was measured by enzymatic assay and Western blotting from blood drawn at the beginning and end of each infusion. Serum ECPKA levels demonstrated a trend to decline with the treatment. In addition to single agent schedules, combination trials were undertaken to assess safety and possible interaction of GEM 231 with taxanes (paclitaxel, docetaxel), given once every 3 weeks (one cycle). While trials using the 2-h twice-weekly GEM 231 infusions are ongoing, preliminary results from both studies show that it is safe to combine paclitaxel or docetaxel with GEM 231. Overall, it is also feasible to administer GEM 231 in combination with taxane or nontaxane chemotherapy (e.g., camptothecins). Phase I combination studies are currently underway to further explore the clinical, pharmacokinetic, and biologic profile of GEM 231 with chemotherapy.

Plant toxic proteins and their current significance for warfare and medicine

Abrin, ricin, viscumin, modeccin, and volkensin are very potent toxins derived from plants. They are glycoproteins composed of two polypeptide chains linked by a disulphide bridge. The A-chain is the enzymatic toxic moiety and B-chain is responsible for bonding to the target cell and internalization of toxin. The toxic part of the toxin molecule removes an adenine from a specific adenosine residue in ribosomal RNA and block proteosynthesis. That is the reason of extreme toxicity of these compounds and their capacity to be used as biological warfare agents or terrorist weapon. Therefore all these compounds are in the schedules of controlled biological agents and toxins. Contrariwise, plant ribosome-inactivating proteins are studied intensive as possible chemotherapeutic agents.

Prevention of resistance: a goal for dose selection for antimicrobial agents.

Drug-resistant microorganisms have become a major problem around the world. In nosocomial and community settings, many important pathogens have demonstrated high-grade resistance to many of our most important agents. In addition, the adverse impact of resistance has not been limited to the bacterial realm. In chemotherapy to treat human immunodeficiency virus (HIV) and other viral diseases, resistance has become a major problem. We are starting to see the beginnings of a resistance problem, even among fungi. Strangely, little attention has been focused on the impact of dosing on the probability with which emergence of resistance occurs. After delineation of the pharmacodynamically linked variable, it is possible to generate dosing regimens that can lower the probability of resistance. In addition, circumstances exist in which combination therapy may be required (e.g., therapy of HIV and tuberculosis). Here, too, it is possible to optimize therapy to prevent resistance by understanding how the drugs in the regimen interact. We can do better with our choices of dose, schedule, and combinations of agents. We will need to lower the probability of resistance and maintain the utility of the drugs currently in our therapeutic armamentarium.

Temporal Reasoning for a Collaborative Planning Agent in a Dynamic Environment

We present a temporal reasoning mechanism for an individual agent situated in a dynamic environment such as the web and collaborating with other agents while interleaving planning and acting. Building a collaborative agent that can flexibly achieve its goals in changing environments requires a blending of real-time computing and AI technologies. Therefore, our mechanism consists of an Artificial Intelligence (AI) planning subsystem and a Real-Time (RT) scheduling subsystem. The AI planning subsystem is based on a model for collaborative planning. The AI planning subsystem generates a partial order plan dynamically. During the planning it sends the RT scheduling subsystem basic actions and time constraints. The RT scheduling subsystem receives the dynamic basic actions set with associated temporal constraints and inserts these actions into the agent's schedule of activities in such a way that the resulting schedule is feasible and satisfies the temporal constraints. Our mechanism allows the agent to construct its individual schedule independently. The mechanism handles various types of temporal constraints arising from individual activities and its collaborators. In contrast to other works on scheduling in planning systems which are either not appropriate for uncertain and dynamic environments or cannot be expanded for use in multi-agent systems, our mechanism enables the individual agent to determine the time of its activities in uncertain situations and to easily integrate its activities with the activities of other agents. We have proved that under certain conditions temporal reasoning mechanism of the AI planning subsystem is sound and complete. We show the results of several experiments on the system. The results demonstrate that interleave planning and acting in our environment is crucial.

The management of the patient undergoing combined modality therapy for locally advanced non-small cell lung cancer.

Multimodality management of locally advanced non-small cell lung cancer is commonplace. Locally advanced disease is defined as patients with stage IIIA, IIIB (without pleural effusions), and Pancoast tumors. Improvements in the median and landmark survival rates of patients have been accomplished by integration of chemotherapy, radiotherapy, and surgery. Although the optimal sequencing and number of modalities remains to be established, the safe treatment of patients requires cooperation of the specialties (joint clinics and tumor boards). The toxicities of chemoradiotherapy are esophagitis, pneumonitis, and myelosuppression. These complications are a function of the dose and fractionation scheme of radiotherapy and the dose, schedule, and specific chemotherapy agents used. The use of modern techniques, including three-dimensional conformal radiotherapy, limitation of field size, and aggressive treatment of symptoms reduces toxicity and the number of treatment breaks, and allows delivery of planned therapy in many patients. In appropriate patients, surgery may be successful after induction chemotherapy or chemoradiotherapy. In addition to exclusion of medically inappropriate patients, patients with persistent mediastinal adenopathy after induction therapy are not likely to benefit from resection. Therefore, mediastinal sampling must be repeated after chemoradiotherapy and before surgery.

Current and future strategies for combined-modality therapy in pancreatic cancer.

Treatment of pancreatic cancer remains a challenging task that often requires a multidisciplinary approach to confer optimal response and, ideally, maximize survival. A combination of locoregional approaches such as surgery and radiotherapy, along with systemic therapies for eradication of micrometastases, should be considered both for patients who are operative candidates and for those with locally advanced, unresectable disease. How best to combine these modalities in terms of schedule, timing, and choice of agents is a question that continues to be actively investigated. Some of these data are equivocal or conflicting; thus standards of care for combined-modality treatment have not been uniformly accepted to date. This article provides an overview of combined-modality therapy, focusing on the major studies that have guided our current approach to the treatment of pancreatic cancer and examining new strategies that are likely to improve outcomes and survival for patients in the future.

Curcumin enhances cytotoxicity of chemotherapeutic agents in prostate cancer cells by inducing p21(WAF1/CIP1) and C/EBPbeta expressions and suppressing NF-kappaB activation.

The modulatory effects and molecular mechanisms of curcumin (CCM) on the cytotoxicity of chemotherapeutic agents to prostate cancer cells were explored. The combined effects of CCM and chemotherapeutic agents were examined by three different administration schedules (one concurrent and two sequential treatments) in two androgen-independent prostate cancer (AIPC) cells (PC-3 and DU145). Alteration of cell cycle progression, protein levels, and transcriptional activation in PC-3 cells were assayed by flow cytometry, Western blotting, and gel shift assay, respectively. The combined effects of CCM --> chemotherapeutic agent schedule showed the greatest synergistic cytotoxicity when compared to the other two schedules in both cells. CCM induced a significant G1 arrest in PC-3, which may be mediated by the induction of p21(WAF1/CIP1) and C/EBPbeta. Moreover, CCM was able to inhibit both the constitutional and TNF-alpha-induced NF-kappaB activation in a time-dependent manner. The incorporation of CCM into cytotoxic therapies may be a promising strategy for the treatment of AIPC.

Temporal Reasoning in Workflow Systems

In a workflow system, autonomous agents perform various activities cooperatively to complete a common task. Successful completion of the task often depends on correct synchronization and scheduling of agents' activities. It would greatly enhance the capabilities of current workflow systems if quantitative temporal constraints on the duration of activities and their synchronization requirements can be specified and reasoned about. This paper investigates such requirements and related reasoning algorithms. In particular, the paper studies the consistency, prediction and enactment services in a workflow system, and provides corresponding algorithms. The consistency service is to ensure that the specification of the temporal constraints is possible to satisfys the prediction service is to foretell the time frame for the involved activitiess and the enactment service is to schedule the activities so that, as long as each agent starts and finishes its task within the specified time period, the overall constraints will always be satisfied. For the enactment service, the paper identifies two practically interesting families of enactment schedules for autonomous agents, namely “free schedules” and “restricted due-time schedules”. In a free schedule, an agent may use any amount of time to finish the task as long as it is between the minimum and maximum time declared by the agent when the workflow is designed. A restricted due-time schedule is a more restrictive one in which the maximum amount of time that an agent may use is limited to a smaller number than the declared maximum. The paper presents efficient algorithms to find free and restricted due-time schedules. The paper also provides algorithms for the above services when multiple time granularities are involved in the temporal constraint specification.