Dodo6/Topic_Modelling_using_LDA
1
1logistics2Article3 4A Novel Integrated FUCOM-MARCOS Model for5Evaluation of Human Resources in a6Transport Company7Željko Stević 1, *819210 11*12 13and Nikola Brković 214 15Faculty of Transport and Traffic Engineering, University of East Sarajevo, Vojvode Mišića 52, 74000 Doboj,16Bosnia and Herzegovina17Transportation company Actros d.o.o, Gorpnji Štrpci bb, 78439 Prnjavor, Bosnia and Herzegovina;18nikolabrkovic@hotmail.com19Correspondence: zeljkostevic88@yahoo.com or zeljko.stevic@sf.ues.rs.ba20 21Received: 22 January 2020; Accepted: 11 February 2020; Published: 13 February 202022 232425 26Abstract: The application of different evaluation approaches in logistics requires considering many27factors with different significance for making the final decision. Multi-criteria decision-making28(MCDM) methods are often applied in logistics to create different strategies and evaluations. In this29paper, research has been carried out in a transport system of an international transport company.30An MCDM model has been created for the purpose of human resource evaluation, on which the31overall efficiency of the company depends. A total of 23 drivers were evaluated on the basis of32five crucial criteria in order to increase employees’ motivation through their periodic remuneration.33The Full Consistency Method (FUCOM) was applied to determine the significance of the criteria, while34the evaluation of potential solutions was performed using Measurement Alternatives and Ranking35according to COmpromise Solution (MARCOS). After the results had been obtained, the created36model was validated throughout comparisons with seven other MCDM methods.37Keywords: logistics; FUCOM; MARCOS; transport; human resource evaluation38 391. Introduction40In order to achieve efficient business performance, it is primarily necessary to ensure a balance41between the needs, requirements, and expectations of the users of a particular service. It is necessary to42make constant measurements and strive to determine the planned set of values that need to be achieved.43Logistics, which provide answers to efficiency and optimization issues, taking into account the44indicators that most influence rationalization, play a very important role in all this. The rationalization45of logistics activities and processes, according to Stević et al. [1], is of utmost importance in the46functioning and the fulfillment of the set goals of every company. This is also confirmed in the study [2]47that emphasizes that the rationalization of primary logistics subsystems plays a key role in achieving48the efficiency and effectiveness of companies.49On the one hand, the transport subsystem of logistics represents the largest percentage of logistic50costs, while on the other hand, it enables achieving the purpose and objectives of logistics, as it affects51the economic system of every country. In order to rationalize the logistic costs incurred by performing52various transport activities, good management that will define adequate strategies is required. Micro,53small, and medium-sized enterprises are suitable and interesting when considering the adoption of54strategies for business process management. In this regard, human resource evaluation is important55since it influences defining business processes and structuring that, according to Dobrosavljević and56Urošević [3], influence the development of business activities of organizations as well as represents a57 58Logistics 2020, 4, 4; doi:10.3390/logistics401000459 60www.mdpi.com/journal/logistics61 62 Logistics 2020, 4, 463 642 of 1465 66starting point on the path to establishing an organization for mature processes. According to Gürbüz67and Albayrak [4], human performance evaluation is one of the most important areas for analyzing the68continuity of an enterprise.69When it comes to developing countries, companies have recognized and accepted the importance70of logistics and have begun to pay more attention to this area over the last few years. This is71especially important for transport companies, as it is necessary to take into account the fleet structure,72human resources, and managers in the subsystems. Thus, there is a need for timely and high-quality73decision-making that positively affects the efficiency of transport companies, since they occur with the74role of a logistics provider obliged to ensure the required quality of service at the lowest possible cost.75The fulfillment of the above is dependent on the age of the fleet, drivers who operate the vehicles, and76decision-makers. In companies that provide logistic services, as is the case in this paper, a large share77of costs is fuel consumption influenced by drivers knowingly or unknowingly. One way to reduce78fuel consumption is to introduce a reward system for drivers who achieve the best results in terms of79fuel consumption.80In this paper, a study has been conducted at a company primarily engaged in international81transport. It is noted that there are significant differences in fuel consumption depending on drivers82individually, even referring to the same transport routes. Therefore, a driver evaluation model has83been created in this study. The most important goal in this paper is the formation of a new model that84should be used for the human resource management in a transport company in order to minimize85costs and increase productivity. The model involves constant driver performance evaluation and most86often, on a monthly basis, the introduction of additional financial reword in order to increase drivers’87motivation. The second goal is the integration of the new MCDM model, which includes The Full88Consistency Method (FUCOM) and Measurement Alternatives and Ranking according to Compromise89Solution (MARCOS) methods, both developed by the first author of this study.90The proper evaluation and selection of personnel in logistics is a very important factor for91optimization, since according to Klumpp and Abidi [5], one of the most important tasks in logistics in92the future is the selection of employees in accordance with their competence and level of knowledge.93Chang [6] created an integrated MCDM model for evaluating employees in a logistics company based94on four criteria: potential for the future, corporate business achievement, organizational commitment,95and working ability. In the paper [7], various investment strategies in the field of human resources96in a logistics company have been evaluated using system dynamics modeling. The research has also97examined the impact of employees on the logistics performance index. Kampf and Ližbetinová [8]98used the AHP method in their research to identify talents in logistics since they saw it as a way to99increase competitiveness. The goal is that after identifying such human resources, they bring new100values to the company. Similar research was carried out in [9–11], where it was observed that the results101obtained could serve as a reference for companies and provide long-term development strategies.102Kucharčíková and Mičiak [12] carried out research in the field of logistic distribution and explained103how human resources and investment in that area could contribute to improving the enterprise’s104own performance and increase its competitiveness in the market. The aim of creating such models,105according to Yue [13], is to understand the degree of correspondence between the demands of the106workplace and the employees in those workplaces, as well as the opportunity to obtain a clearer picture107of human resources in a logistics company.108After the introductory part presenting the significance of the research, motives for its realization,109goals, and a brief review of the situation in the field, the paper is divided into the following sections.110The second section presents the methods used in this study and their steps: FUCOM and MARCOS. The111third section is a case study that involves the evaluation of driver performance. A sensitivity analysis112through the comparison of the applied model with other methods, and calculation of Spearman’s113correlation coefficient (SCC) is provided in the fourth section. In addition, the obtained results are114discussed and compared with other methods in this section. The fifth section includes the conclusion115with the contributions of the study and directions for future research.116 117 Logistics 2020, 4, x1184 FOR PEER REVIEW119 12033of121of 14122 1232. Methods1242. Methods1252.1. Full Consistency Method: FUCOM1262.1. Full Consistency Method: FUCOM127The FUCOM method is based on the principles of pairwise comparison and validation of128The FUCOM method is based on the principles of pairwise comparison and validation of results129results through deviation from maximum consistency [14]. Benefits that are determinative for the130through deviation from maximum consistency [14]. Benefits that are determinative for the application131application of FUCOM are a small number of pairwise comparisons of criteria (only n − 1132of FUCOM are a small number of pairwise comparisons of criteria (only n − 1 comparison), the ability133comparison), the ability to validate the results by defining the deviation from maximum consistency134to validate the results by defining the deviation from maximum consistency (DMC) of comparison and135(DMC) of comparison and appreciating transitivity in pairwise comparisons of criteria. The FUCOM136appreciating transitivity in pairwise comparisons of criteria. The FUCOM model also has a subjective137model also has a subjective influence of a decision-maker on the final values of the weights of138influence of a decision-maker on the final values of the weights of criteria. This particularly refers to139criteria. This particularly refers to the first and second steps of FUCOM in which decision-makers140the first and second steps of FUCOM in which decision-makers rank the criteria according to their141rank the criteria according to their personal preferences and perform pairwise comparisons of142personal preferences and perform pairwise comparisons of ranked criteria. However, unlike other143ranked criteria. However, unlike other subjective models, FUCOM has shown minor deviations in144subjective models, FUCOM has shown minor deviations in the obtained values of the weights of145the obtained values of the weights of criteria from optimal values [14–17]. Additionally, the146criteria from optimal values [14–17]. Additionally, the methodological procedure of FUCOM eliminates147methodological procedure of FUCOM eliminates the problem of redundancy of pairwise148the problem of redundancy of pairwise comparisons of criteria, which exists in some subjective models149comparisons of criteria, which exists in some subjective models for determining the weights of150for determining the weights of criteria. Figure 1 presents the FUCOM algorithm [18].151criteria. Figure 1 presents the FUCOM algorithm [18].152 153Algorithm: FUCOM154Input: Expert pairwise comparison of criteria155Output: Optimal values of the weight coefficients of criteria/sub-criteria156Step 1: Expert ranking of criteria/sub-criteria.157Step 2: Determining the vectors of the comparative significance of evaluation criteria.158Step 3: Defining the restrictions of a non-linear optimization model.159Restriction 1: The ratio of the weight coefficients of criteria is equal to the comparative160significance among the observed criteria, i.e. wk wk +1 = ϕk /(k +1) .161Restriction 2: The values of weight coefficients should satisfy the condition of162mathematical transitivity, i.e. ϕk /(k +1) ⊗ϕ(k +1)/(k +2) = ϕk /(k +2) .163Step 4: Defining a model for determining the final values of the weight coefficients of evaluation164criteria:165min χ166st. .167wj (k )168wj (k +1)169wj(k )170wj (k +2)171 172−ϕk /(k +1) ≤ χ, ∀j173−ϕk /(k +1) ⊗ϕ(k +1)/(k +2) ≤ χ, ∀j174 175n176 177w = 1178j179 180j =1181 182wj ≥ 0, ∀j183 184Step 5: Calculating the final values of evaluation criteria/sub-criteria ( w1, w2 ,..., wn ) .185T186 187Figure 1. FUCOM188FUCOM methodology189methodology [18].190 1912.2. Measurement192Measurement Alternatives193Alternatives and194and Ranking195Ranking According196According to197to COmpromise198Compromise Solution1992.2.200Solution(MARCOS)201(MARCOS)202In this203presented.204The205MARCOS206method207is based208on209In210this section,211section, the212thealgorithm213algorithmofofMARCOS214MARCOSmethod215methodisis216presented.217The218MARCOS219method220is based221defining222the relationship223between224alternatives225and reference226values (ideal227and anti-ideal228alternatives).229on230defining231the relationship232between233alternatives234and reference235values236(ideal and237anti-ideal238On239the240basis241of242the243defined244relationships,245the246utility247functions248of249alternatives250are251determined,252alternatives). On the basis of the defined relationships, the utility functions of alternatives and253are254compromise and255ranking256is made ranking257in relation258to ideal259anti-ideal260solutions.261Decision262preferences263are264determined,265compromise266is made267in and268relation269to ideal270and anti-ideal271solutions.272Decision273defined274on275the276basis277of278utility279functions.280Utility281functions282represent283the284position285of286an287alternative288with289preferences are defined on the basis of utility functions. Utility functions represent the position of an290regard to anwith291idealregard292and anti-ideal293solution.294The bestsolution.295alternative296the one297that is closest298thethat299ideal300alternative301to an ideal302and anti-ideal303Theis best304alternative305is the to306one307is308and309at310the311same312time313furthest314from315the316anti-ideal317reference318point.319The320MARCOS321method322is323performed324closest to the ideal and at the same time furthest from the anti-ideal reference point. The MARCOS325through is326the327followingthrough328steps [19]:329method330performed331the following steps [19]:332 333 Logistics 2020, 4, 4334 3354 of 14336 337Step 1: Formation of an initial decision-making matrix. Multi-criteria models include the definition338of a set of n criteria and m alternatives. In the case of group decision-making, a set of r experts should339be formed to evaluate alternatives according to the criteria. In the case of group decision-making,340expert evaluation matrices are aggregated into an initial group decision-making matrix.341Step 2: Formation of an extended initial matrix. In this step, the extension of the initial matrix is342performed by defining the ideal (AI) and anti-ideal (AAI) solution.343 344X=345 346AAI347A1348A2349...350Am351AI352 353C1354355x356357aa1358359 x11360361 x36236321364365 . . .366367 x368369m1370371xai1372 373...374...375...376...377...378...379...380 381C2382xaa2383x12384x22385...386x22387xai2388 389Cn390xaan391x1n392x2n393...394xmn395xain396 397398399400401402403404405406407408409410 411(1)412 413The anti-ideal solution (AAI) is the worst alternative, while the ideal solution (AI) is an alternative414with the best characteristic. Depending on the nature of the criteria, AAI and AI are defined by applying415Equations (2) and (3):416AAI = min xij i f j ∈ B and max xij i f j ∈ C417(2)418i419 420i421 422AI = max xij i f j ∈ B and min xij i f j ∈ C423i424 425i426 427(3)428 429where B represents a benefit group of criteria, while C represents a group of cost criteria.430Step431h i 3: Normalization of the extended initial matrix (X). The elements of the normalized matrix432N = nij433are obtained by applying Equations (4) and (5):434m×n435 436nij =437 438nij =439 440xai441if j ∈ C442xij443xij444 445if j ∈ B446 447xai448 449(4)450 451(5)452 453where elements xij and xai represent the elements of the matrix X.454h i455Step 4: Determination of the weighted matrix V = vij456. The weighted matrix V is obtained by457m×n458multiplying the normalized matrix N with the weight coefficients of the criterion w j , Equation (6).459vij = nij × w j460 461(6)462 463Step 5: Calculation of the utility degree of alternatives Ki. By applying Equations (7) and (8), the464utility degrees of an alternative in relation to the anti-ideal and ideal solution are calculated.465Ki− =466 467Si468Saai469 470(7)471 472Ki+ =473 474Si475Sai476 477(8)478 479where Si (i = 1,2,..,m) represents the sum of the elements of the weighted matrix V, Equation (9).480Si =481 482n483X484i=1485 486vij487 488(9)489 490 Logistics 2020, 4, 4491 4925 of 14493 494Step 6: Determination of the utility function of alternatives f(Ki ). The utility function is the495compromise of the observed alternative in relation to the ideal and anti-ideal solution. The utility496function of alternatives is defined by Equation (10).497Ki+ + Ki−498 499f (Ki ) =5001+501 5021− f (Ki+ )503f (Ki+ )504 505+506 5071− f (Ki− )508 509;510 511(10)512 513f (Ki− )514 515 516 517where f Ki− represents the utility function in relation to the anti-ideal solution, while f Ki+ represents518the utility function in relation to the ideal solution.519Utility functions in relation to the ideal and anti-ideal solution are determined by applying520Equations (11) and (12).521 522Ki+523−524f Ki = +525(11)526Ki + Ki−527 528f Ki+ =529 530Ki−531Ki+ + Ki−532 533(12)534 535Step 7: Ranking the alternatives. This is based on the final values of utility functions. It is desirable536that an alternative has the highest possible value of the utility function.5373. Case Study538The research was carried out at an international transport company. Due to the high criteria539and professional attitude of the analyzed transport company, the fleet consists of the most modern540transport units that today’s market has to offer. Such a fleet structure results in low maintenance541costs since the vehicles are extremely young, lower toll costs due to cheaper tariffs for vehicles with542the highest Euro standards, and lower costs of consumables such as oil, petroleum, and AdBlue.543Due to the large number of clients of this company, besides trucks, this company possesses the most544modern and extremely young trailers for goods that require transport under tarpaulin or thermal545mode, semi-trailers equipped for ADR shipments, as well as semi-trailers intended for the transport546of working machines with the ability to move poles for their easier loading or unloading. Despite547these advantages, segmentation and diagnostics of the whole system have been performed in order to548identify potential structures for improvement.549Throughout the detailed research within the company, improvement activities have been defined550for each subsystem. In order to create a set of actions to improve the logistics performance of551the company, human resource management has been identified as a potential part of the business552performance that can be significantly improved. It should be emphasized that currently, there is quite553a good system of employee motivation. However, the management of human resources, i.e., drivers of554freight road vehicles, is a potential area for increasing the efficiency of the company since the adequate555performance of the transport service is practically most dependent on their engagement. It has been556found that there is a need to evaluate the performance of drivers periodically and to reward two to557three best drivers on a monthly basis. In accordance with that, it may be noted that there is a need to558develop such an integrated multi-criteria approach.5593.1. Forming a Multi-Criteria Decision-Making Model560In order to evaluate the performance of drivers adequately and to solve problems, the criteria for561human resource evaluation have been defined. The established set of criteria consists of a combination562of quantitative and qualitative criteria, and there are five criteria in total. List of criteria for evaluation563has adopted in cooperation with manager in the company according to their needs. The first criterion564(C1) is fuel consumption, which is expressed in liters per 100 km. The criterion belongs to the cost group565and it should be minimized. It also belongs to the group of quantitative criteria. This information566 567 Logistics 2020, 4, 4568 5696 of 14570 571is available from the database maintained on a daily basis for each vehicle, for each driver and for572kilometers traveled. The second criterion (C2) is damage per kilometer and is reflected in the monetary573expression of the damage to the vehicle caused by the driver’s carelessness and negligent behavior.574The criterion covers all damages regardless of the amount of damage and it is of quantitative and575cost type. The third, fourth, and fifth criteria need to be maximized and they are of a qualitative576type, i.e., they are determined on a basis of linguistic scale. The third criterion (C3) refers to the577adequacy of vehicle maintenance by drivers, which is periodically controlled by the manager. The578fourth criterion (C4) refers to the driver’s ability to provide information timely and adequately. This579criterion primarily refers to the driver’s ability to communicate adequately and timely with a dispatcher.580This is very important in certain unforeseen circumstances, which are relatively common situations581in a transport process. The fifth criterion (C5) is loyalty, which means performing all the obligations582without complaint and therefore the flexibility of drivers to perform the tasks.5833.2. Determining the Weights of Criteria Using the FUCOM Method584In order to be able to evaluate the performance of drivers, it is necessary to determine the585significance of the criteria first. For this purpose, the FUCOM method is applied. In the first step,586the ranking of criteria according to the real needs of the company, i.e., C1 > C2 > C4 > C3 > C5 is587determined. Subsequently, based on the preference of decision-makers who made it by consensus,588the criteria are compared by applying the scale 1–9, which is shown in Table 1.589Table 1. Comparisons of criteria.590Criteria591 592C1593 594C2595 596C4597 598C3599 600C5601 602Comparisons603 6041605 6061.7607 6083.4609 6104.5611 6125613 614The procedure for determining the significance of the criteria is as follows:615First, the comparative priority of the criteria is determined:616ϕC1 /C2 = 1.7/1 = 1.7; ϕC2 /C4 = 3.4/1.7 = 2; ϕC4 /C3 = 4.5/3.4 = 1.32; ϕC3 /C5 = 5/4.5 = 1.11617In the next step, the final values of the weight coefficients are calculated, and they should meet618the two conditions:619Condition (1):620w1 /w2 = 1.7; w2 /w4 = 2; w4 /w3 = 1.32; w3 /w5 = 1.11;621and Condition (2):622ϕC1 /C4 = 1.7 ∗ 2 = 3.4; ϕC2 /C3 = 2 ∗ 1.32 = 2.64; ϕC4 /C5 = 1.32 ∗ 1.11 = 1.47623Thence:624w1 /w4 = 3.4; w2 /w3 = 2.64; w4 /w5 = 1.47625The final model from which the final criteria values are obtained is presented:626minχ627 w628w4629w26301631632633634w2 − 1.7 = χ, w4 − 2 = χ, w3 − 1.32 = χ,635636637638w4639w26401641642 w643w4 − 1.7 = χ, w3 − 2.64 = χ, w5 − 1.47 χ644s.t.6456466475648649P650651652w j = 1, w j ≥ 0, ∀j653654655j=1656 657w3658w5659 660− 1.11661 662 663w4664w3665w5666 w2667668w669w670w671s.t. 1 − 1.7 = χ , 2 − 2.64 = χ , 4 − 1.47 χ672w3673w5674 w4675 5676 w j = 1, w j ≥ 0, ∀j677 j =1678 679Logistics 2020, 4, 4680 6817 of 14682 683Subsequently,684Subsequently, using685using the686the Lingo687Lingo software688software (17689(17 version),690version), the691the following692following model693model is694is obtained695obtained with696with the697the698weights699of700the701criteria702shown703in704Figure7052.706weights of the criteria shown in Figure 2.707 708Figure7092. Results710weights.711Figure 2.712Results of713of applying714applying the715the FUCOM716FUCOM method717method –– criterion718criterion weights.719 720can be721beseen722seenininFigure723Figure724most725important726is first727the first728criterion,729fuel consumption,730As can7312, 2,732thethe733most734important735is the736criterion,737fuel consumption,738with awith739valuea740value741of7420.434,743which744is745understandable746since747the748most749important750improvements751in752savings753are754of 0.434, which is understandable since the most important improvements in savings are expected755expected756from this This757criterion.758This isby759followed760by the761second,762costthat763criterion,764relates765to the766damage767from768this criterion.769is followed770the second,771cost772criterion,773relates that774to the775damage776caused777and778caused779and of780has0.128.781a value7820.128.three783The criteria,784other three785criteria,786are of a type,787qualitative788type, have789much790has791a value792Theofother793which794are ofwhich795a qualitative796have much797less values798less values799than800first801two and802therefore803much less significance.804than805the first806twothe807and808therefore809much810less significance.8113.3. Evaluation of Human Resources Using the MARCOS Method812The initial matrix, which has been extended in accordance with the second step of the MARCOS813method, i.e., Equations (1)–(3), is shown in Table 2. The first and second criteria are of cost type and814according to Equation (2), the anti-ideal solution (AAI) is the highest characteristic, which is 35.7 for815the first and816values817ofof8185,819and 0.1008200.100 for821for the822thesecond823secondcriterion.824criterion.For825Forthe826thethird,827third,fourth,828fourth,and829andfifth830fifthcriteria,831criteria,the832theleast833least834values83555,and8363, respectively,837areare838included839in the840AAIAAI841solution.842Applying843Equation844(3), the845included846into8475 and8483, respectively,849included850in the851solution.852Applying853Equation854(3), values855the values856included857the858solution859(AI) are860For the861criteria,862thosethose863are the864values865of 25.8866and867intoideal868the ideal869solution870(AI)determined.871are determined.872Forcost873the cost874criteria,875aresmallest876the smallest877values878of 25.88790.001,880while881for the882criteria,883the highest884values885are included886in AI,887which888is aisvalue889of 9offor890all891and 0.001,892while893for benefit894the benefit895criteria,896the highest897values898are included899in AI,900which901a value9029 for903the904criteria.905It is906toto907note908that909all the910criteria.911It important912is important913note914thata avalue915valueofof0.0019160.001ininthe917theextended918extendedinitial919initialmatrix920matrix indicates921indicates that922the drivers did not make any damage, but that indicator is used for the purpose of calculation and to923avoid a value of zero.924After expanding the initial matrix, normalization should be performed by applying Equation (4)925for cost and Equation (5) for benefit criteria. An example of the normalization shown in Table 3 is as926follows:927for cost criteria: nij =928 929xai930xij931 932for benefit criteria: nij =933 934i f j ∈ C ⇒ n11 =935xij936xai937 93825.893933.9940 941i f j ∈ B ⇒ n13 =942 94379449945 946= 0.761947= 0.778948 949 Logistics 2020, 4, 4950 9518 of 14952 953Table 2. Extended initial decision matrix.954Criteria955 956C1957 958C2959 960C3961 962C4963 964C5965 966AAI967A1968A2969A3970A4971A5972A6973A7974A8975A9976A10977A11978A12979A13980A14981A15982A16983A17984A18985A19986A20987A21988A22989A23990AI991 99235.799333.999430.099531.599633.099731.999831.199934.8100025.8100131.7100235.7100334.2100429.6100530.2100634.3100731.0100834.5100933.9101033.1101132.7101235.0101331.8101431.0101532.5101625.81017 10180.10010190.00110200.00110210.00110220.00110230.00110240.05010250.00110260.00110270.00110280.00110290.00110300.10010310.00110320.00110330.00110340.00110350.00110360.00110370.00110380.00110390.00110400.00110410.00110420.0011043 104451045710469104791048710499105071051710527105371054710557105671057910587105951060710615106291063910649106551066910677106891069 107051071910727107371074910757107671077710787107971080710817108271083510847108571086710877108891089910909109171092710937109491095 109631097710987109991100911017110291103711047110591106711079110871109511109111171112711137111491115911167111791118711193112091121 1122Table 3. Normalized decision matrix.1123Criteria1124 1125C11126 1127C21128 1129C31130 1131C41132 1133C51134 1135AAI1136A11137A21138A31139A41140A51141A61142A71143A81144A91145A101146A111147A121148A131149A141150A151151A161152A171153A181154A191155A201156A211157A221158A231159AI1160 11610.72311620.76111630.86011640.81911650.78211660.80911670.83011680.74111691.00011700.81411710.72311720.75411730.87211740.85611750.75211760.83211770.74811780.76111790.77911800.78911810.73711820.81111830.83211840.79411851.0001186 11870.01011881.00011891.00011901.00011911.00011921.00011930.02011941.00011951.00011961.00011971.00011981.00011990.01012001.000