Resource Allocation Management Research Articles

Treatment of chronic kidney disease in older populations.

As the world population ages, an expected increase in the prevalence of chronic kidney disease (CKD) among older individuals will pose a considerable challenge for health care systems in terms of resource allocation for disease management. Treatment strategies for older patients with CKD should ideally align with those applied to the general population, focusing on minimizing cardiovascular events and reducing the risk of progression to kidney failure. Emerging therapies, such as SGLT-2 inhibitors and GLP-1 receptor agonists, hold promise for the effective management of CKD in older individuals. In addition, non-pharmacological interventions such as nutritional and exercise therapies have a crucial role. These interventions enhance the effects of pharmacotherapy and, importantly, contribute to the maintenance of cognitive function and overall quality of life. Various factors beyond age and cognitive function must be taken into account when considering kidney replacement therapy for patients with kidney failure. Importantly, all treatment options, including dialysis, transplantation and conservative management approaches, should be tailored to the individual through patient-centred decision-making. The dynamic integration of digital technologies into medical practice has the potential to transform the management of CKD in the aging population.

Key Factors Affecting the Success of Construction Multi Project Management

Simultaneous management of multiple projects has become a trend in the field of energy, oil and gas construction project management over the past decade. This situation is driven by the high market demand and need for fossil fuels, which also stresses project management discipline to improve its performance and efficiency in managing multiple projects simultaneously. Nevertheless, concurrent execution of multiple construction projects often leads to stretched the project durations and cost due to suboptimal resource management, which may become a boomerang and increase project costs. The objective of this study is to identify factors and indicators that contribute to the successful management of multiple projects. Although multi-project management has been extensively studied, there is still a dearth of studies analyzing the elements that make construction multi-project management successful, particularly in the energy, oil, and gas sectors. This research is a case study of an EPCIC company that uses multi-project management and operates in the oil and gas sector. Through purposive sampling techniques, a total of 58 executives from national oil and gas contractors with significant work experience participated in this study through a questionnaire survey. The PLS-SEM approach has revealed that the positive predictor factors influencing the success/performance of multi-project management are resource allocation, teamwork quality, multi-project management competency, and project manager assignment. Flexible and intelligent resource allocation management in multi-project cycles affected by uncertainty is a critical factor.

When to monitor or control: Informed invasive species management using a partially observable Markov decision process (POMDP) framework

Abstract Resource allocation for invasive species management requires information about the size of the invasive population, which may be expensive and time‐consuming to obtain. The trade‐off between investment in monitoring and control efforts is a challenging decision problem, and existing mathematical tools are often difficult to interpret, and/or limited to a specific case study. We propose a partially observable Markov decision process (POMDP) framework to help decision‐makers understand effective monitoring and control policymaking. POMDPs can deal with uncertainty in both the model and state of the system but are more challenging to solve due to the continuous and high‐dimensional state space. Rather than limiting the possible states of the system, as do most previously proposed methods, we work through the development of a density projection approach, which reduces the dimensionality of the space of beliefs by restricting them to a parametrised family of probability distributions. This serves to align the mathematical representation of the problem with the real‐world quantities relevant to human decision‐making. The result of our model is a sequence of actions, which minimises the expected cost incurred in managing the invasive species, where the recommendation depends on an estimate of the species' abundance, and the uncertainty in this estimate. We demonstrate the effectiveness of our proposed framework with a case study on tropical fire ant (Solenopsis geminata) control and two generic case studies of varying complexity. Furthermore, we investigate sensitivity of the results to the choices of control cost and efficacy, and monitoring cost and error. The framework proposed by this paper makes the powerful machinery of POMDPs available to environmental managers. It computes the optimal course of action to manage a growing population of an invasive species, incorporating a varying time horizon and multiple control interventions. We sidestep the computational difficulties of general POMDPs to provide a clear, visual overview of decision‐making recommendations, and how these decisions change in new situations. Initial results and scenario‐based analysis show promising results, and the framework could be extended to the related field of disease management.

Leveraging graph neural networks for supporting automatic triage of patients

Patient triage is crucial in emergency departments, ensuring timely and appropriate care based on correctly evaluating the emergency grade of patient conditions. Triage methods are generally performed by human operator based on her own experience and information that are gathered from the patient management process. Thus, it is a process that can generate errors in emergency-level associations. Recently, Traditional triage methods heavily rely on human decisions, which can be subjective and prone to errors. A growing interest has recently been focused on leveraging artificial intelligence (AI) to develop algorithms to maximize information gathering and minimize errors in patient triage processing. We define and implement an AI-based module to manage patients’ emergency code assignments in emergency departments. It uses historical data from the emergency department to train the medical decision-making process. Data containing relevant patient information, such as vital signs, symptoms, and medical history, accurately classify patients into triage categories. Experimental results demonstrate that the proposed algorithm achieved high accuracy outperforming traditional triage methods. By using the proposed method, we claim that healthcare professionals can predict severity index to guide patient management processing and resource allocation.

Evaluation of a rapid diagnostic test for detection of Vibrio cholerae O1 in the Democratic Republic of the Congo: Preventative intervention for cholera for 7 days (PICHA7 program).

Globally, there are estimated to be 2.9 million cholera cases annually. Early detection of cholera outbreaks is crucial for resource allocation for case management and for targeted interventions to be delivered to stop the spread of cholera. In resource limited settings such as Eastern Democratic Republic of the Congo (DRC), there is often limited laboratory capacity for analysing stool samples for cholera by bacterial culture. Therefore, rapid diagnostic tests (RDTs) for cholera present a promising tool to rapidly test stool samples in a health facility setting for cholera. Our objective is to evaluate the Crystal VC O1 RDT for cholera detection compared with bacterial culture and polymerase chain reaction (PCR) for Vibrio cholerae. From March 2020 to December 2022, stool samples were collected from 644 diarrhoea patients admitted to 94 health facilities in Bukavu in Eastern DRC. Patient stool samples were analysed by Crystal VC O1 RDT for cholera and by bacterial culture and PCR for V. cholerae O1. Twenty six percent of diarrhoea patients (166/644) had stool samples positive for cholera by RDT, and 24% (152/644) had stool samples positive for V. cholerae O1 by bacterial culture or PCR. The overall specificity and sensitivity of the Crystal VC O1 RDT by direct testing was 94% (95% confidence interval [CI]: 92%-96%) and 90% (95% CI, 84%-94%), respectively, when compared with either a positive result by bacterial culture or PCR. Our findings suggest that the Crystal VC O1 RDT presents a promising tool for cholera surveillance in this cholera endemic setting in sub-Saharan Africa.

Admission Prediction Using Time Series Analysis

Efficient resource allocation and strategic planning in educational institutions heavily rely on accurate predictions of student admissions. This paper presents a detailed investigation into the application of time series analysis techniques for admission prediction. We explore the utilization of historical admission data, spanning multiple academic cycles, to train and evaluate several time series models.The study encompasses a comparative analysis of traditional statistical models such as autoregressive integrated moving average (ARIMA) and seasonal autoregressive integrated moving average (SARIMA) alongside more advanced deep learning techniques like long short-term memory (LSTM) networks. By evaluating the performance metrics, including accuracy, robustness, and computational efficiency, we aim to identify the most suitable model for admission prediction tasks. Moreover, the research delves into the impact of various external factors such as changes in the academic calendar, socio-economic indicators, and demographic shifts on admission patterns. Understanding these factors is crucial for enhancing the predictive capabilities of the models and enabling institutions to adapt their strategies accordingly. The experimental results and comparative analysis provide valuable insights for educational institutions, enabling them to make data-driven decisions regarding enrollment management strategies and resource allocation. Ultimately, this research contributes to the advancement of admission prediction methodologies, facilitating more efficient and informed decision-making processes in the educational domain.

Business Process Reengineering with a Circular Economy PDCA Model from the Perspective of Manufacturing Industry

In times of increasing awareness of sustainability and the need for efficient business processes, this study explores the integration of business process reengineering with circular economy principles within Serbian manufacturing organizations. Addressing the need for sustainable development, the research aims to propose and validate a model that harmonizes business process reengineering with the circular economy to improve environmental and organizational performance. The study conducted an extensive survey and analysis across 135 manufacturing organizations in Serbia, assessing their readiness and current practices in adopting circular economy strategies through business process reengineering, utilizing the Plan-Do-Check-Act (PDCA) model. The findings reveal a moderate level of integration, with an average implementation score of 44.70% across surveyed organizations. Notably, organizations with ISO 9001 and ISO 14001 certifications demonstrated higher levels of model implementation. The study highlights the potential of integrating business process reengineering with circular economy principles as a path to sustainable manufacturing. It also highlights the need for targeted strategies to improve management commitment, resource allocation, and participation in sustainable practices. The research contributes valuable insights for policymakers, industry stakeholders, and academic discourse, advocating for a more systematic approach to embedding circular economy principles within organizational processes for a sustainable future.

Research on Improving the Optimization System of Agricultural and Forestry Resource Allocation Management Measures by Using Multi-Objective Optimization Algorithm

At present, there are some problems in the use of forestry resource database, such as large amount of data and data centralization, which leads to the low efficiency of the system. Therefore, an application software platform for forestry informatization is built based on optimized Hadoop cluster. The system consists of four parts: permission management, file management, interface management and data processing. In this paper, a control strategy FDMDR is designed, which implements dynamic copy management according to user access frequency. A method of file access frequency limit based on DMDR is proposed, and the data in the system is processed dynamically. The experiment proves that the system can integrate traditional files, images, videos and other contents organically. The system has achieved a better integration of traditional agricultural information resources. It has good performance and scalability.

A Statistical Forecasting Model for Extremes of the Fire Behaviour Index in Australia

The increasing frequency and duration of severe fire events in Australia further necessitate accurate and timely forecasting to mitigate their consequences. This study evaluated the performance of two distinct approaches to forecasting extreme fire danger at two- to three-week lead times for the period 2003 to 2017: the official Australian climate simulation dynamical model and a statistical model based on climate drivers. We employed linear logistic regression to develop the statistical model, assessing the influence of individual climate drivers using single linear regression. The performance of both models was evaluated through case studies of three significant extreme fire events in Australia: the Canberra (2003), Black Saturday (2009), and Pinery (2015) fires. The results revealed that ACCESS-S2 generally underestimated the spatial extent of all three extreme FBI events, but with accuracy scores ranging from 0.66 to 0.86 across the case studies. Conversely, the statistical model tended to overpredict the area affected by extreme FBI, with high false alarm ratios between 0.44 and 0.66. However, the statistical model demonstrated higher probability of detection scores, ranging from 0.57 to 0.87 compared with 0.03 to 0.57 for the dynamic model. These findings highlight the complementary strengths and limitations of both forecasting approaches. Integrating dynamical and statistical models with transparent communication of their uncertainties could potentially improve accuracy and reduce false alarms. This can be achieved through hybrid forecasting, combined with visual inspection and comparison between the statistical and dynamical forecasts. Hybrid forecasting also has the potential to increase forecast lead times to up to several months, ultimately aiding in decision-making and resource allocation for fire management.

D2D Matching Techniques for Resource Allocation Management and Power Control Management in B5G/6G

The upcoming 6G network is expected to have a great integration of Device-to-Device (D2D) technology. One of the key advantages of D2D technology is its potential to minimize the load on cellular base stations, which can help extend the lifespan of cellular network infrastructure. D2D multicast communication can meet the growing demand for multimedia content in local area services by maximizing energy efficiency and device lifespan by reusing the same resources for the cellular network. In this paper, we investigate an efficient resource allocation scheme by using the One-To-Many Gale Shapley pairing algorithm (GSA) for efficient allocation and power optimization. We propose a joint optimization approach for D2MD clusters that considers the Signal-to-Interference Noise ratio and energy levels of devices' batteries. The problem is optimized, so it is divided into two convex sub-problems. In the first sub-problem, power allocation is performed for each candidate cluster-head (CH)and cellular user to maximize energy efficiency using the Dinkelbach matching algorithm. In the second sub-problem, the One-to-Many Gale Shapley matching algorithm is used to optimize resource allocation and cluster-head selection to select the Cluster Nodes (CRns) to form the cluster. Numerous investigations show that the suggested technique maintains QoS and minimal battery power requirements while increasing cluster-head longevity and energy efficiency (EE) in D2D applications.

OPTIMIZING FMCG SUPPLY CHAIN DYNAMICS: A NOVEL FRAMEWORK FOR INTEGRATING OPERATIONAL EFFICIENCY AND CUSTOMER SATISFACTION

This paper proposes a groundbreaking theoretical model specifically designed for the FMCG sector, focusing on optimizing supply chain dynamics to balance operational efficiency with customer satisfaction. Drawing on hands-on strategies and problem-solving abilities, the framework aims to offer practical solutions for common challenges faced by FMCG operations managers, such as inventory management, demand forecasting, and efficient resource allocation. The fast-moving consumer goods (FMCG) industry is characterized by intense competition, short product life cycles, and rapidly changing consumer preferences. In this challenging environment, optimizing supply chain dynamics is crucial for FMCG companies to achieve operational efficiency and enhance customer satisfaction. This concept paper presents a novel framework for integrating operational efficiency and customer satisfaction in FMCG supply chains, aiming to provide insights and strategies for FMCG companies to improve their supply chain performance. The framework emphasizes the importance of aligning supply chain strategies with customer needs and preferences. By focusing on key areas such as inventory management, distribution network efficiency, and technology integration, FMCG companies can improve their supply chain agility and responsiveness. Additionally, the framework highlights the importance of leveraging data analytics and technology to enhance supply chain visibility and decision-making. Implementing the framework requires collaboration among stakeholders, including suppliers, manufacturers, distributors, and retailers. By working together to optimize supply chain processes and enhance customer service, FMCG companies can gain a competitive advantage in the marketplace. Overall, the framework offers a comprehensive approach to optimizing FMCG supply chain dynamics, with the ultimate goal of improving operational efficiency, reducing costs, and enhancing customer satisfaction.
 Keywords: FMCG, Supply Chain, Customers, Framework.

ПРОЄКТУВАННЯ БАГАТОПРОДУКТОВИХ ТРАНСПОРТНИХ МЕРЕЖ З ДИСКРЕТНИМИ ПОТОКАМИ МАГІСТРАЛЬНИХ ДАННИХ

The article considers the importance of transport and communication as key sectors for the development of the economy of Ukraine, as well as the need to create applied mathematical software for the optimization of transport processes and the development of information and analytical systems. Special attention is paid to the development and study of multi-product communication networks with discrete flows of small consignments, which have a significant impact on market transformations in Ukraine, especially in the field of transportation of small consignments. The study focuses on modern and future communication networks, which, as a rule, have a multi-level structure determined by the administrative division of the territory, management hierarchy and technologies for the processing and distribution of goods and information. The authors of the article emphasize the need to design multi-level communication networks, which would take into account the complexity of the system, as well as the impossibility of complete centralization of information processing and management in one link. The article also focuses on the analysis of existing research and publications in this field, paying attention to the contribution of foreign and domestic scientists to the development of models and methods of network resource management. Most research focuses on deterministic models and models based on the mathematical foundations of mass service theories and random Markov processes. The authors indicate the inadequacy of the existing mathematical tools for modeling and analyzing the functioning of multi-product networks with discrete flows, which requires the development of new mathematical models and methods.The purpose of the research is the development and analysis of effective approaches to the design of multi-tracks pipeline transport networks with a special focus on discrete trunk data flows. The authors note that this includes the development of new approaches and methods for effective management and optimization of resource allocation, as well as modeling the specifics of discrete flows in the context of trunk transport systems. An important part of the research is the creation of a hierarchical system structure for automated resource management and flow distribution in large-scale networks. The authors emphasize the need to develop a methodology that integrates already existing achievements in this field and allows to effectively solve practical problems related to long-term development, ongoing planning and operational management in large-scale networks. Such a methodology should include models of different levels of aggregation and be suitable for all levels of the network’s hierarchical structure. The study also includes consideration of mathematical models and algorithms for modeling and analyzing the functioning of multi-product networks with discrete flows.

Energy Poverty Impact on Sustainable Development of Water Resources in China: The Study of an Entropy Recycling Dynamic Two-Stage SBM Model

Water scarcity is increasingly being recognized as a global concern. Sustainable Development Goal 6 (SDG-6) was established by the United Nations to address water resource governance within its sustainable development framework. This study employs the entropy weight method and a two-stage cyclical dynamic slacks-based measure (SBM) model to scientifically evaluate water resource utilization and governance across various regions in China. The findings reveal notable disparities in both the production and governance efficiency of water resources. Recognizing governance efficiency is crucial for promoting sustainable water resource utilization and socioeconomic development. The eastern, central, and western regions encounter unique challenges in attaining sustainability. The eastern region exhibits minimal potential for enhancing technical efficiency, necessitating interventions in management strategies and resource allocation. Conversely, the challenges in the central and western regions are more pronounced, demanding immediate implementation of new technologies and equipment. The data analysis in this study yields conclusions that offer targeted improvement recommendations to address disparities across China’s eastern, central, and western regions, and this is achieved by considering various developmental stages and regional contexts. These recommendations cover areas such as technical support, financial investment, and policy incentives, with the aim of enhancing the sustainable utilization of water resources in the country.

Hospitalisation expenditure on tuberculosis among tribal populations in India: A repeated cross-sectional analysis of national sample survey data, 2004 to 2018

ObjectiveTribal population in India (8.6% of the total population) have a greater prevalence of tuberculosis compared to the national average. The article aims to study out-of-pocket expenditure (OOPE), hardship financing, and impoverishment effects of TB hospitalisation treatment among tribal populations in India. MethodsData of three rounds of National Sample Surveys (NSS) 60th (2004–05), 71st (2013–14) and 75th (2017–18) rounds were analyzed. Descriptive statistics, bivariate estimates and multivariate models were performed to calculate the OOPE, healthcare burden (HCB), catastrophic health expenditure (CHE), hardship financing and impoverishment effects using standard definitions at February 2023 price values. Propensity score matching (PSM) was used to examine the effect of health insurance coverage on catastrophic health expenditure, and impoverishment. ResultsOver two-thirds of the TB cases are seen in the economically productive age group (14–59 years). Substantial OOPE and its impact on HCB, CHE, and poverty impact observed among 15–35 age group across all three rounds. Illiterate patients and those availing private hospitals for TB treatment had higher OOPE, HCB, hardship financing, CHE, and poverty impact. 38.5% (2014) and 33.2% (2018) are covered with any kind of public healthcare coverage, PSM analysis shows households with health insurance have lower incidence of CHE and impoverishment effects due to TB hospitalisation expenditure. ConclusionsThe current study aids in comprehending the patterns in the financial burden of TB on tribal households during the previous 15 years and gives policy makers information for efficient resource allocation management for TB among Indian tribal communities.

Expert Consensus on Improving Stroke Care Ecosystem in India

The stroke burden in India is alarmingly high compared to Western industrialised countries, necessitating immediate action to develop stroke systems of care. Tailored stroke-prevention strategies, adequate resource allocation for stroke management and research, and increased awareness of specialised care in acute Stroke are critical. The COVID-19 pandemic has underscored the need for improved healthcare delivery and infrastructure. To address the rising stroke burden in India, comprehensive strategies encompassing prevention, awareness, specialised care and governmental support are urgently required. While the concept of ‘stroke-ready centres’ has emerged, their limited availability hinders widespread benefits. This expert consensus examines the current state of stroke care in India, identifies challenges, and proposes strategies to enhance stroke care systems. It offers valuable insights into stroke management at different levels, empowering healthcare professionals and administrators to assess their existing systems and make necessary improvements. Special attention is given to the role of physicians, including the use of thrombolytic therapies like Tenecteplase, in reducing the stroke burden and improving outcomes across the country. Significant strides can be made in stroke management and care in India by addressing these critical areas.

Research on Cloud Computing

Abstract: Cloud computing represents a pivotal advancement in the IT sector, offering efficient management and distribution of vast amounts of data and resources across the internet. Essentially, it enables accessing IT infrastructure over a network without local installations. Businesses benefit from the flexibility to adjust resource levels according to operational demands, leading to reduced infrastructure costs. Additionally, cloud computing facilitates faster application testing, streamlined management, and adaptable resource allocation. Its widespread application across various contexts underscores its significance in daily life. This paper comprehensively examines cloud computing, encompassing its architecture, characteristics, varieties, service models, advantages, and challenges.

Influencing Factors and Evaluation of Groundwater Ecological Function in Arid/Semiarid Regions of China: A Review

In arid and semi-arid areas, due to drought climate and shortage of water resources, groundwater is crucial for natural ecological protection and economic development. It serves a dual role as a resource function and an ecological function. However, with the continuous improvement of the exploitation and utilization of groundwater by human activities during rapid economic development, the phenomenon of groundwater overexploitation is becoming more and more serious, which has destroyed the natural balance of groundwater recharge and discharge. As a result, natural vegetation has lost the maintenance of the ecological function of groundwater, and a series of ecological and environmental problems have occurred, such as natural vegetation degradation, land desertification, sandstorms, and so on. In recent years, scholars have carried out research on groundwater resource management and optimization of water resource allocation, trying to solve the problem of water balance in arid regions. However, there is still a lack of comprehensive understanding and systematization regarding influencing factors and degeneration mechanisms related to groundwater’s ecological function. By summarizing and analyzing the previous research results, this paper summarizes the influencing factors, evaluation methods, existing problems and future directions of groundwater ecological function research in China to provide a reference for rational exploitation and utilization of groundwater and ecological protection. This paper is divided into four main contents. The first part introduces the definition of groundwater ecological function (GEF); the second part summarizes the research status of influencing factors of GEF, including the groundwater table depth, vegetation root system and lithologic structure of vadose zone, etc.; the third part analyzes the evaluation of groundwater ecological function; the fourth part discusses the existing problems in the study of groundwater ecological functions, and based on the above research the evaluation framework of GEF is proposed with the Shiyang River basin as a case study; and finally, it highlights the future research directions about GEF.

The role and significance of operations research in management

The article explores the pivotal role of Operations Research (OR) in contemporary management, tracing its origins from World War II to its evolution into an indispensable tool for optimizing decision-making in various industries. OR, characterized by mathematical modelling, statistical analysis, and optimization techniques, aligns seamlessly with management objectives, enhancing efficiency, resource allocation, and strategic decision-making. The multifaceted contributions of OR are elucidated, encompassing its role in strategic decision-making, the development of Decision Support Systems, resource allocation, risk management, and interdisciplinary collaboration. Emphasizing the dynamic and multifaceted nature of OR, the article underscores its significance in shaping effective management strategies. The core tenet of optimization becomes a catalyst for efficiency gains and strategic resource utilization. As organizations grapple with the complexities of a rapidly evolving business landscape, the integration of OR principles is portrayed not merely as beneficial but imperative for sustained success. It emerges as a guiding philosophy, empowering organizations to navigate uncertainties, make informed decisions, and foster innovation and excellence. The synergy between Operations Research and management is positioned as a strategic imperative, steering organizations towards efficiency, innovation, and long-term prosperity in an era where data-driven decision-making is paramount.

Behavior Prediction of Vespa mandarinia based on Convolutional Neural Networks

Vespa mandarinia poses a significant threat to honey bees and beekeepers. This paper aims to accurately predict the presence and behavior of Vespa mandarinia to enable effective control strategies. We develop an "Asian hornet queen prediction model" to forecast the dispersal path and number of queens, identifying their concentration in Bellingham, Washington, and Bellingham, Canada. Using a convolutional neural network, we achieve 96.13% accuracy in identifying Vespa mandarinia images. Analyzing confidence levels of unprocessed and unverified labels reveals a significant number of high-confidence samples. Incorporating human control factors into the model, we find 60% human intervention to be most effective in reducing the number of Asian hornet queens. Ultimately, our research highlights the potential for anthropological or ecological measures to eliminate Vespa mandarinia populations, aiding in resource allocation for proactive management.

Multi-Compression Scale DNN Inference Acceleration based on Cloud-Edge-End Collaboration

Edge intelligence has emerged as a promising paradigm to accelerate DNN inference by model partitioning, which is particularly useful for intelligent scenarios that demand high accuracy and low latency. However, the dynamic nature of the edge environment and the diversity of end devices pose a significant challenge for DNN model partitioning strategies. Meanwhile, limited resources of the edge server make it difficult to manage resource allocation efficiently among multiple devices. In addition, most of the existing studies disregard the different service requirements of the DNN inference tasks, such as its high accuracy-sensitive or high latency-sensitive. To address these challenges, we propose a Multi-Compression Scale DNN Inference Acceleration (MCIA) based on cloud-edge-end collaboration. We model this problem as a mixed-integer multi-dimensional optimization problem, jointly optimizing the DNN model version choice, the partitioning choice, and the allocation of computational and bandwidth resources to maximize the tradeoff between inference accuracy and latency depending on the property of the tasks. Initially, we train multiple versions of DNN inference models with different compression scales in the cloud, and deploy them to end devices and edge server. Next, a deep reinforcement learning-based algorithm is developed for joint decision making of adaptive collaborative inference and resource allocation based on the current multi-compression scale models and the task property. Experimental results show that MCIA can adapt to heterogeneous devices and dynamic networks, and has superior performance compared with other methods.