دکتر کریم آتشگر

One of the most practical methods of monitoring and controlling project performance is earned value management. The widespread use of this method in many projects leads to the production of multiple indicators with mutual influence, This important feature makes their individual analysis with errors. In this method, not only the mutual influence of the indicators is not paid attention to, but also the lack of attention to the monitoring of the variability of the indicators has caused this method to monitor the project based on the criterion of indexability. In this research, using multivariable control charts, we have simultaneously checked the performance indicators of a gas supply project based on real data. After analyzing the obtained results, we have compared the performance of Hotelling and (MEWMA) charts with each other. After comparison, it was found that MEWMA chart has more capability and sensitivity than Hotelling chart in identifying changes in multivariate processes.

In the current business environment with ever-increasing changes, innovation is a necessity for firms to survive. In this regard, technology-oriented companies spend a lot of resources on research and development and acquisition of new technologies with the aim of achieving technological innovation. One of the main factors determining the success of firms in this field is the adoption of appropriate and aligned strategies at the business and operational levels. The current research aims to investigate the strategy of Iran's pioneering medical equipment manufacturing companies in achieving technological innovation, and investigates the business, technology, and research and development strategies in the pioneering medical equipment manufacturing companies and the relationship between these strategies and the situation The environmental factors affecting this adoption of the technology strategy of the companies including the technological and market regimes and the innovation policies of the government have been examined. For this purpose, three pioneering medical equipment manufacturing firms were studied using the multiple case study method, and deep and semi-structured interviews were conducted with the CEOs and research and development managers of these firms and 4 officials in the field of medical equipment policy making. Results of research show that the adopted strategies of these firms are completely aligned. Also, the most important environmental factors affecting the decisions of companies in adopting strategy have been discussed.

Supplier cooperation strategy plays an important role in the strength of the organization in today's market. The ability to monitor supplier performance and monitor the cooperation strategy between the organization and the supplier is a critical capability to maintain a strong organization-supplier relationship. This article examines the selection of the cooperation model with the supplier and monitoring the cooperation strategy over time by continuously monitoring the performance of the supplier over time. In this paper, a multi-variable multi-profile monitoring model is presented in two stages for monitoring the supplier's unanticipated costs. In the first step, criteria and  indicators to evaluate the delivery process, quality are extracted according to library resources and the opinion of experts and elites of the industry. In the next phase, in the first phase of monitoring, using the T^2 method, model parameters based on successive differences are            calculated and monitored, and in the second phase, the likelihood ratio method is used to monitor the supplier's unforeseen costs over time. to show its control chart in the shortest possible time if there is a warning. In the last stage, by using the results of the profile of the delivery process and the quality and monitoring of the unforeseen costs, a decision is made regarding the selection and monitoring of the cooperation strategy with the supplier. This model has been implemented in the automotive industry of Iran Khodro Company and the supplier of gearbox parts by Niro Travezah Company.

This article investigates the selection of a cooperation model with suppliers and the continuous monitoring of the collaboration strategy over time by tracking the supplier's performance. A multiple fuzzy profile monitoring model is proposed and implemented in two stages to oversee the supplier’s unforeseen costs. In the first stage, criteria and indicators for evaluating the delivery process and quality are identified based on literature reviews and expert opinions from industry leaders. In the second stage, a two-phase monitoring approach is adopted. First, using the T² method, the model’s parameters based on successive differences are calculated and separately monitored for the multiple fuzzy profiles of delivery and quality processes. In the second phase, the likelihood ratio method is applied to track the profiles over time, enabling the control chart to signal any warning in the shortest possible time. In the final stage, the results from the fuzzy profiles of delivery and quality processes, combined with artificial intelligence and the fuzzy inference system tool, are used to monitor unforeseen costs, make decisions regarding the supplier, and assess the adopted strategy. This model has been implemented in Iran's automotive industry, specifically within Iran Khodro Company and its gearbox parts supplier, Niromoharkeh Company.

Today, degradation analysis is one of the most important approaches in evaluating the reliability of multi-component systems. As it is clear, improving the performance of real systems requires the use of efficient and predictable approaches to analyze degradation with considering the interaction of system degradation processes on each other. The literature review shows that degradation analysis of multi-component systems has been investigated in various researches, but the approach in which there is a profile relationship between the degradation processes of the system components has not been considered so far. When there is a functional relationship between the degradation processes of one or more components, it is called a profile in the statistical process control literature. The aim of this study is to provide an efficient approach to predict and evaluate the variability of degradation processes in the presence of multivariate profiles under the conditions of stochastic dependence. In fact, the proposed approach offers the possibility of predicting and evaluating the variability of degradation processes at the component and system level. In this paper, in order to evaluate the proposed approach, the data set of a multi-component system with a structure of 2 out of 3 has been used. The results show the effectiveness of the proposed approach.

The ability to monitor supplier performance is a critical capability to maintain a strong organization-supplier relationship, especially for organizations where the supplier plays a significant role in their end product and market retention. In this article, a multi-variable multi-profile monitoring model is presented, taking into account the risk conditions in two stages for monitoring the quality of parts and supplier products. In the first step, the criteria and indicators for evaluating the quality process as well as the risk function are extracted according to the library resources and the opinion of experts and elites of the relevant industry. In the next phase, in the first phase of monitoring, using the Hotelling’s  Statistic Based on the Successive Differences, the parameters of the model are calculated and monitored, and in the second phase, the likelihood ratio method is used to monitor the supplier's quality process over time, so that if the presence of a warning will show the control chart in the shortest possible time. This model has been implemented in the automotive industry of Iran Khodro Company and the supplier of gearbox parts by Niro Travezah Company.

Compressive strength is one of the important properties of concrete, which is affected by factors such as humidity, water-cement ratio, cement composition, aggregate dimensions, and age of concrete. All cases except age can be controlled by making changes in the mixing plan, therefore, in this research, the relationship between age and compressive strength of concrete has been investigated using profile analysis in a 6-story building construction project by concrete quality control experts. In this research, 10 concrete samples were processed in laboratory conditions after preparation, and the compressive strength of the samples was checked by 3 experts and each expert checked each sample 3 times. In the zero phase, with the Gage study method, the data is not affected by the measurement error. In phase one, by using the Storr and Braille control chart, out-of-control profiles have been removed, and a model has been presented to estimate the compressive strength of concrete according to the age of the sample. In the second phase, using 10,000 data simulated by MATLAB software, the performance of the model has been examined using the (EWMA) chart, the results show that the EWMA chart is able to identify out-of-control signals by applying changes in the model parameters An out-of-control signal can be a warning for signs such as cracking, cracking, sanding, and discoloration at older ages of concrete. Therefore, managers and concrete quality control experts can predict the factors of concrete destruction and take measures to restore concrete without spending a lot of money.

The ability to monitor supplier performance is a critical capability to maintain a strong organization-supplier relationship, especially for organizations where supplier quality plays a significant role in their final product and market retention. In this article, a profile monitoring model in two phases is presented for monitoring the quality of products and supplier processes. In the first step, criteria and indicators for quality assessment are extracted according to library resources and the opinion of experts and elites of the industry. In the next phase, in the first phase of monitoring, the parameters of the model are calculated and monitored using the Kong Albain T^2 method, and in the second phase, the Hotling T^2 method is used to monitor the quality of the supplier over time, so that if there is a warning in the shortest time The control chart may show it. This model has been implemented in the automotive industry of Iran Khodro Company and the supplier of gearbox parts by Niromoharkeh Company.

This paper aims to analyze comprehensively and structurally the different methods proposed in the literature by researchers, and identify the new fields for future studies.

Design/methodology/approach: 

In this research, the methods proposed in the literature have been explained comprehensively and structurally focusing on mathematical functions, assumptions, functional objectives, results, weaknesses, and strengths. Different types of dependencies between system components have been also considered in this study.

In this paper, the methods in the literature were clustered by a new approach based on the system type, dependency, the method of considering the dependency and the objective. The different challenges of this field of study in the future were also investigated in this paper.

Research limitations/implications:

 Few practical studies in the field of multi-component systems have been performed in the literature. 

Practical implications: 

Since the analysis of multi-component systems is considered an important and applied issue in the reliability field for different complex products, the discussion of this paper and the results can be used by technical experts in different industries.

Social implications:

 This paper can be used as a reference for increasing people’s quality of life.

Originality/value: 

To the best of the authors’ knowledge, this is the first time that a comprehensive analysis of multi-component systems is reported in considering its structure and applied approach

Tracking military structures and equipment is one of the parameters to create superiority in military battles. Camouflage has long been used to reduce the possibility of detection of military structures and equipment. Development of knowledge in the field of tracking and discovering military structures and equipment followed by the necessity of using the concealment in order to reduce vulnerability in war, has enhanced the importance of using new camouflage and radar evasion methods. The use of nanotechnology in the field of radar evasion of military facilities was developed by introducing graphene as a very strong absorber of electromagnetic waves. Graphene coating on the military installations causes the absorption of electromagnetic waves and as a result, these installations are not detected by the enemy's radar.  Referring to the fact that there is a functional relationship between the diameter of the graphene oxide used and the radar evasion of the equipment, an attempt has been made in this article for the first time to find a solution to control and monitor the radar evasion quality using graphene oxide (with the scientific name of RGO/NiFe2O4) in the microwave spectrum of 7GHZ by the profile approach and presentation of a regression relationship. This model can be used to monitor the quality of radar evasion products (cheaper and faster than existing methods). Finally, sensitivity analysis of the model showed that the ability to detect non-conformity in the manufactured products can be detected quickly (between 1 and 20 samples) with the change in the parameters of the regression model.

One of the first steps taken in the epidemic of a disease such as Covid-19 in the community is to conduct an epidemiological study. In this case, the basic reproduction number (R0) index is referred to as a key term to analyze and control the pandemic condition. The fluctuation of this index really is affected by different factors. The objective of this paper is: 1) to design a functional relationship between R0 term and the temperature variable, and 2) to illustrate that the pandemic condition of a community can be monitored effectively by the profile approach. The Covid-19 virus was first detected in Wuhan, China, and has infected more than 35 million people as of this writing. In the case of Covid-19 virus, the term pandemic is used instead of epidemic after the virus has spread worldwide and has been in such a way that it has crossed several continents and has affected all countries of the world (1). One of the first measures to control the prevalence of such diseases is to conduct an epidemiological study. A key feature in epidemiological assessment and control is the Basic Reproduction Number ( ).    means that each carrier can infect an average of several others by entering an unprotected population. In epidemiology, there are several factors affecting   (2). One of the most important uncontrollable factors whose effects on the Covid-19 virus pandemic have been discussed by many researchers so far, is environment temperature (5).

The method used in this research to provide the objectives is profile monitoring. Profile monitoring is categorized in statistical approach. In this research the functional relationship (profile) between temperature variable and the basic reproduction number is analyzed. Experimental analysis methods are also used leading to achieve the desired functional relationship. This paper, for the first time, uses profile monitoring to offer a new approach to pandemic control and management while proposing a functional relationship between environment temperature and Basic Reproduction Number. Monitoring this relationship is important in that it can indicate the occurrence of uncontrolled conditions in this relationship. Out-of-control conditions in the functional relationship between environment temperature and   mean that the variability of the pandemic functional model has changed under the influence of certain factor or factors that have not been considered by the authorities. This paper, while proving the existence of a functional relationship between the Basic Reproduction Number and the environment temperature, seeks to show that pandemic can be effectively monitored with a profile monitoring approach. In other words, a process is controlled using profile monitoring approach, and by designing a functional model. Profile monitoring is a knowledge that goes beyond production systems today and is used in other fields such as engineering, environment, biotechnology, genetics, medicine and finance (7). This research has also used experimental analysis methods to achieve the desired functional model. The data required for the research were gathered from two valid sources. The first part of this data is related to the daily statistics of corona cases over a period of 60 days from June 18, 2020 to August 16, 2020 for 10 countries with the most affected (United States, Brazil, India, Russia, South Africa, Mexico, Peru, Chile, Spain and Iran). The second part is data about the weather conditions and average temperature parameters of the 10 countries in the mentioned period.

In this research not only a profile provided based on basic reproduction number and temperature variables, but also it addressed that a pandemic condition of the community can be monitored by a profile relationship effectively. Furthermore this research showed that the variability due to abnormal factors and natural variability can be identified by the profile approach.Based on the experimental analysis, the studied countries were divided into two subsets based on the similarity of the adopted non-pharmaceutical policies. The first group includes Brazil, India, Spain, Peru and South Africa, and the second group includes Iran, the United States, Russia, Chile and Mexico. Fisher's Least Significant Difference (LSD) were used to select a subset of countries whose policies had similar results on the Basic Reproduction Number. Before starting the profile design and monitoring phase, it was necessary to evaluate all the data used in the research. The first part of this evaluation was related to the Basic Reproduction Number data in the studied countries in the desired time period. In this section, the normality and variance of data were evaluated and confirmed. Also, to prove the suitability of profile models, it should be concluded by performing various statistical tests that the errors of the uncorrelated random variable mean is equal to zero and variance is also equal. These errors also should follow the normal distribution (8). This was also done in the data analysis section. As a result, in order to analyze the model according to the effect of the policies adopted by governments, the research data for each group of countries studied were analyzed separately. This article examines the proposed model in two phases and analyzes the sensitivity of the proposed profile monitoring model for three different time periods in Iran. The results of this analysis show that the designed control model can monitor the pandemic conditions of Covid-19 virus well, affected by changes in temperature and show good signals of the occurrence of factors affecting the Basic Reproduction Number.

This paper investigated the proposed model for two phases (that is phase-I and phase-II) and analyzed the sensitivity of the proposed profile monitoring for three different time periods in Iran. The results of this analysis showed that the designed control model is capable of monitoring effectively the pandemic conditions of Qovid-19 virus using the functional relationship between an independent variable (that is temperature) and a response variable (i.e. R0) and it is capable of producing appropriate signals.Examining the functional relationship between Basic Reproduction Number and other factors affecting this important index, and continuous monitoring of this functional relationship, can provide pandemic control and management in any country with more accurate and in-depth information. So that it is possible to trace the root causes of the decrease or increase of the pandemic more effectively. In this study, the profile monitoring method was used to monitor and control the functional relationship between the Basic Reproduction Number and environment temperature. Using experimental design method and regression analysis, this paper presented a profile model for Covid-19 pandemic monitoring for Iran and several other countries. Sensitivity analysis of the proposed model and the use of three real-time timelines showed that this model could yield good results for Covid-19 pandemic managers. Obviously, this new approach can have a significant impact on the management of the Covid-19 pandemic.

This research proposes a new model for customer experience management and investigates the impact of customer experience management on improving the performance of one of the Iranian banks, mediated by customer satisfaction, loyalty and verbal interaction. This research develops a customer experience management model for the banking process, and then to analyze the structural equation modeling, data gathering is conducted in the field. In this study, in addition to collecting the necessary data from customers, the real performance data of the bank is provided by the performance evaluation section of the bank branches. The information includes deposit performance, receivables, electronic services and financial performance of the branches. This research is applied in terms of purpose and is descriptive-correlative in terms of data collection. This research not only provides statistical analysis of the model quality, but also addresses the results of impact analysis of customer experience management on performance indices, costomer satisfaction, and costomer loyalty. The results of this study shows that the use of the proposed customer experience management model not only is capable of improving the core competency of the studied bank, but also the approach can lead the bank to improve its performance terms.

The prevalence of Covid-19 virus is referred to as one of the most important issues in the world. Providing an effective solution to control or treat these patients type can be of great help to human society. The purpose of this paper is to provide a method for monitoring effectively the condition of patients with Covid-19 virus approaching a functional relationship called profile.

In this paper, for controlling and monitoring the condition of patients with Covid-19 a solution is proposed. The existence of relationship between CD4 T cells and lymphocyte count provided the opportunity of designing a profile to monitor the condition of the patient. The information of the patients published in scientific sources is used to analyze the validation of the proposed model.

Although in the process of treatment of patients with Covid-19, the number of CD4 T cells and the number of lymphocytes are considered separately, this paper proposes that these two parameters should be considered in a functional relationship at the same time. The findings of this paper show that considering the proposed functional relationship provides a more effectiveness condition for a physician’s judgment.

The results of this research show that the profile approach provides an effective condition to analyze the status of a patient with Covid-19 considering a functional relationship between CD4 T and lymphocyte count. In addition, the results of this research are capable of leading one to use the proposed approach for evaluating vaccines and drugs prepared for this disease type.

Literature indicates that the statistical process control approach is capable of improving the quality of surgical processes. The use of this approach in the case of real surgical processes is an attractive contribution for researchers. Since risk factors of each patient before a surgery are different potentially, the use of a control chart approaching risk-adjusted is addressed by researchers.The sensitivity of a model to small shift types of the mortality odds ratio is very important for monitoring the quality of a surgery process. Although different schemes have been proposed by researchers to monitor the performance of a medical team, based on the authors' knowledge this research for the first time proposes a risk-adjusted double exponentially weighted moving average (RADEWMA) approach to monitor the mortality rate. The numerical performance analysis of this paper based on average run length (ARL) term for the cardiac surgery addresses that the sensitivity of the proposed risk-adjusted scheme is superior compared to other risk-adjusted chart of literature.In this research the performance of the proposed model is compared comprehensively with risk adjusted P chart (RAP), risk adjusted exponentially weighted moving average (RAEWMA), and risk adjusted cumulative sum (RACUSUM). In this paper for analyzing the performance comparatively, the simulated data is produced based on UK Center for Cardiac Surgery data in the period of 1992-1994. This data is referred to 2218 patients with heart surgery. In this simulation to estimate the parameters of the logistic regression, Parsonnet number is used as an explanatory variable. The comparative performance analysis is reported for different shift cases. The comparative report is analyzed statistically for different values of parameters. For statistical analyzing SPSS software is also used. The new proposed scheme to monitor the mortality odds ratio, can be used in different surgeries of a hospital effectively.

In recent years, due to the limitation of fossil fuels and the environmental impacts of using such fuels, focusing on renewable energy sources has significantly increased. In developed countries, using clean energy such as wind power has been considered as an alternative source. Monitoring the performance of wind turbines and controlling their output power quality is one of the important issues for managing wind farm. One of the influential characteristics of a wind turbine is the power curve which depicts the relationship between output power and hub height wind speed. Therefore, accurate models of power curves can play an important role in improving the performance of wind-energy-based systems. Hence, this paper has carried out the parametric techniques for power curve modeling. First, the following parametric equations are introduced to represent the power curves of wind turbines: polynomial power curve, exponential power curve, cubic power curve and approximate cubic power curve. Next, the models have been applied to a wind turbine, and, then, they were compared with manufacturer’s normal power curves by using the goodness of fit test. The results have shown that the exponential method for modeling of power curve is more desirable compared with three other models.

Identification of the change point in panel data leads practitioners to focus on the time when really a change takes place in the cross sectional data. Identification of the time helps one to provide a more realistic analysis of the change manifested itself to the process. Different methods of change point identification have been proposed in literature, however, the literature addressees that the sensitivity of identifying the change point is an important issue. This paper attempts to propose a new method with high sensitivity for identifying the change point in a panel data (with large dimension) through a hybrid approach. The proposed method is named Double CUSUM-EWMA. The comparative report addresses that the performance of the proposed method has relatively better performance compared to the existing methods in the literature. This study analyzes several simulated numerical examples with large dimension of panel data when a step shift manifests itself to the process.

Introduction & Breast cancer is known as the most common site-specific cancer as well as it is referred to as the second cause of mortality for women. Literature indicates that the most important prognostic factors are addressed by lymph node involvement, the activity of estrogen and progesterone receptors, tumor size, tumor grade, histologic type, and epidermal growth factor. Published researches address that these prognostic factors and the relationship between them are known as important indicators to assess the effectiveness of diagnosis of a treatment method and a treatment method influence of the breast cancer. However the published researches do not address analyzing the impact of the receptor status on lymph node involvement and the impact of the relationship between the prognostic factors on survival of patients with breast cancer. This study attempts to not only investigating the impact of the receptors on lymph node involvement, but also the survival of breast cancer patients considering simultaneously the receptor status and lymph node involvement prognostics, is analyzed.
In this study, 1822 patients diagnosed with breast cancer have been considered. These patients after treating in the period of 2007 till mid 2016 were cared under adjuvant therapy such as hormone therapy, chemotherapy and radiotherapy. In this study, receptor groups are defined in 8 categories based on different combinations, as well as the number of lymph nodes involvement are categorized in 4 branches (i.e. no lymph node involvement, between 1-3 lymph nodes involvement, between 4-9 lymph nodes involvement, and more than 9 lymph nodes involvement). Receptor statuses of these patients were analyzed using the logistic regression approach. In this research, the statistical software R has been used for data analysis.
The age mean (± Standard deviation) of the patients were addressed by 50.45 (±11.47) years. Lymph node involvement was recognized for more than half of the patients (933 people i.e. 51.20%). The frequency of each 8 receptor groups were obtained as (263 individuals) 14.43%, (222 individuals) 12.18%, (762 individuals) 41.82%, (138 individuals) 7.57%, (40 individuals) 2.19%, (70 individuals) 3.84%, (11 individuals) 0.54%, (22 individuals) 1.20%, respectively. Group 3 of the receptor status (i.e. simultaneous prognostics of positive estrogen, positive progesterone, and negative HER2), addressed the highest percentage of (762 individuals) 41.82%, and Group 7 (i.e. simultaneous prognostics of negative estrogen, positive progesterone, and HER2-positive) addressed the lowest percentage of (11 individuals) 0.54% among the other groups. P-value is addressed by 0.01 and 0.05 for estrogen and HER2, respectively, when the impact of receptors on lymph nodes has been analyzed.
This study indicated that: 1) there exist relationship between estrogen receptors and Her2 with lymph node involvement, so that these receptors impact on involving the lymph nodes. 2) In the case that the prognostics of positive estrogen, positive progesterone, and negative Her2 occur simultaneously, the receptor status addresses the best condition of the survival for a patient with breast cancer compared to other groups.

Published reports indicate that only a limited prognostic factor of breast cancer has been addressed by researchers. In other words, the relationship of all the known prognostic factors corresponding to breast cancer patients is not addressed by literature. Analyzing the relationship is capable of affecting the process of treating breast cancer patients. This paper focuses on studying comprehensively the relationship between all prognostic factors corresponding to breast cancer women.
This cross-sectional survey used the data of Sheikh Aliyan etal. (1). In this study, 1822 patients with breast cancer, who in 1395-1386 years were treated in three hospitals of Tehran, have been studied. This research used pathological examination, histopathological report, and magnetic imaging to complete patients’ data. In this study, SPSS software is used to analyze data statistically.
This comprehensive study addresses which prognostic factors of breast cancer have relationship with each other significantly. So that, this research indicates that there is a significant relationship for the factors as the follow:A.Between age and prognosis of 1) the type of pathology, 2) lymph node involvement, 3) the number of lymph nodes involved, and 4) the type of surgery
B.Between pathology, chemotherapy and prognostic of 1) lymph node involvement, 2) number of lymph nodes involved, 3) the type of surgery, 4) the type of pathology
C.Between the receptor and the prognosis of 1) grade 2) the size of the tumor, 3) stage of the disease, 4) the type of surgery and 5) metastasis
D.Between the involvement of lymph nodes with involved lymph node nimbers, chemotherapy, and prognostic of 1) grade 2) the size of the tumor, 3) stage of the disease, 4) types of surgeries, 5) metastasis
E.Between the grade and prognostic of 1) the size of the tumor, 2) stage of the disease, 3) the type of surgery and 4) metastasis
F.Between tumor size, chemotherapy and prognosis of 1) stage of the disease, 2) the type of surgery, 3) metastasis
G.Between stage of disease, chemptherapy and the prognosis of 1) the type of surgery, 2) metastasis
H.Between the type of surgery, chemptherpy and the metatstssis prognosis
There is a significant relationship for %50.4 of the known prognosis of breast cancer; however, the relationship for %49.6 of them is not detected significantly.

Introduction & Developing the use of statistical process control (SPC) methods for the healthcare area is evaluated as an attractive contribution. Analyzing patient data using SPC approach plays an important role to monitor effectively a health care process, improving the quality surgical processes, and other medical service processes. Among SPC methods, cumulative sum approach have been attracted more attention in the medical literature due to its simple formulation and interpretation, as well as more sensitivity to detect medium and small shifts compared to other methods. The aim of this study is to provide analytically positive effects of using CUSUM charts for hospitals, surgical, and therapeutic processes.
This study is a comprehensive literature review. This paper investigates comprehensively researches and results of the used CUSUM charts in different areas of healthcare focusing on reported reliable scientific sources without any limitation of time.
This study reveals that different cumulative sum control chart types are proposed to use in therapeutic and surgical processes leading to improvement of the processes. The analysis indicates that risk adjustment methods considering patients’ risk before surgery is more superior compared to other CUSUM types. This approach focuses on homogenizing and adjusting the effect of physical and physiological conditions corresponding to patients.
Cumulative sum charts should be used considering strength and weakness points of each chart as well as data probability distribution. This paper reveals that the use of risk-adjusted cumulative sum charts can be more accurately reflect the performance results of a surgery compared to other chart types. Furthermore the use of CUSUM methods for learning curves is capable of monitoring effectively the performance of medical trainees.