Reducing the Lead Time of Litho Printing Sample Making Process at Avery Dennison Lanka
In:
Submitted By iraz89 Words 5855 Pages 24
PROJECT TITLE Reducing the lead time of litho printing sample making process at Avery Dennison Lanka INTRODUCTION Avery Dennison Lanka (Pvt) Ltd is a company, which is situated in Biyagama free trade zone. Avery Dennison is a multinational company which is spread among 60 countries over the world. The company ranked number 362 on the 2010 fortune 500 list of the largest U.S. industrial and service companies. The CEO is Dean A. Scarborough. There are around 600 employees working in the company. Avery Dennison develops, manufactures and sells products through four groups of businesses as Pressure‐sensitive Materials, Retail Information Services, Office and Consumer Products and other specialty converting businesses. This Company's products include pressure‐sensitive labelling materials, graphics imaging media, retail apparel ticketing and branding systems, RFID inlays and tags, office products, specialty tapes, and a variety of specialized labels for automotive, industrial and durable goods applications. In Sri Lanka, products are narrowed down to labels and tags which give information about a specific garment good. Garment manufacturers can be considered as the main customers of Avery Dennison Lanka (Pvt) Ltd. Thus such tags are to be manufactured in large quantities. In order to get a satisfactory product that meets the customer need, first the samples has to be fabricated. The sample making process can be identified as the bottle‐neck of the whole litho‐printing process. Consuming money, labour and time. This report contains an analytical study of the sampling process and suggestions to make the lead time minimum accordingly. OBJECTIVE(S) Report mainly focuses on the sample production process in the company. SCOPE Among 5 different label production lines we selected lithography production method and we mainly focused on the sample making process of litho printing section. METHODOLOGY Study the existing sample making process and identify weak points which increase the lead time of sample making process. Analyse all the operations and technology involved in making samples in litho printing section. Collecting required data on sample making process, o o o How the front end handle with customers How the design centre works on a design How the current sample making process perform (flow process) To manage the both sample production and the mass production effectively within the limited facilities. To enhance the quality and the accuracy of the sample. To minimize the lead time of sample making process.
o Number of main processes and sub processes _______________________________________________________________________________________D OME Page 2 of 18 ME4042/09/01
o o o o
Number of workers and machines Allocated space Identifying the skills and experience of labourers How they assess the quality of a sample
Perform a time study to identify the time deployment for each process. Research into more efficient options for sample making process which minimize the lead time and enhance the quality. SAMPLE PRINTING PROCESS ANALYSIS
In order to get a better understanding of the sampling process a time and data study was done. What we found was that the bottle necking in the process is so severe, that it cripples down the efficiency and extends the lead time by nearly 40%. Initial lithography process is as followed
CUSTOMER
FRONT DESK ACTIVITIES
DESIGN CENTER JOBS
PRINTING UNIT JOB
1. AN ORDER IS RECIEVED
4. REPLICATING THE SAMPLE
5. COLOUR MATCHING
2. CHECKING FOR EXISTING SAMPLE DATA
NO SEARCH HITS THEN, IF DATA EXISTS THEN,
6. SAMPLE PRINTING
7. CHECKING SAMPLE QUALITY
8. PRINTING & CUTTING
9. DELEVERY TO CUSTOMER
_______________________________________________________________________________________D OME Page 3 of 18 ME4042/09/01
From the data received, what we found was there is a average of 10‐12 days of lead time for a non‐ prioritised job. ACTIVITY/JOB Checking for excising data for samples To receive sample data from the headquarters at Hong Kong Designing/ Prefabricating Colour matching & sample printing NUMBER OF WORKING DAYS(APPROXIMATELY) 1 day 2‐3 days 3days 2days (depending on the colour complexity E.g.: two colour/ four colour ) 1day 2‐3 days(depending on the size of the order) 20% 50% TIME PERCETAGE (100%)TO 10DAYS 10% 20%
Rechecking for flows and errors Printing
In front end handling process they negotiate with customers to obtain the art work and other relevant facts for their print. Here they firstly check whether that the print they get from the customer is in the data base of Avery Dennison in Hong Kong which intercedes between headquarters and Avery Dennison Sri Lanka, if so they download a design kit from Hong Kong data base and forward it into printing section. In case the print given by customer is not in their data case they have to design it by themselves. In the design centre they create plates which have a roughened texture and are coated with a photosensitive (light sensitive) emulsion. This emulsion is a suspension of two chemicals that cannot be mixed together. After the plate design process they are directed to the printing section to print. In the label printing section it has 2 four‐colour machines and 2 two‐colour machines to work on printing for samples as well as for orders. Once the plates are being directed to the printing section they insert plates in to machines and make the exact colours according to the customer request by mixing some basic colours. This mixing of colours is done by manually and they get the exact colours for the print by their experiences. Next they print some samples and check the colours of the print after they are dried and they do this procedure until they get the correct colours of the print. Sometimes they repeat this for three or four times to get the exact colours they need. Once they found the correct combination of colours they start printing the samples for the customer. To achieve this correct combination of colours they have to print about 500 sheets (6000 sample prints). As the last prints are in a good quality they select 30 printed labels for customer to get the approval for bulk production. As observed, designing to printing takes roughly around 5‐6 working days. In this time period lithography machines are idled or forced to work on smaller jobs as the colour matching is done reputedly to get the maximum accuracy. Compared to other time consuming events, sampling is the most costly process. _______________________________________________________________________________________D OME Page 4 of 18 ME4042/09/01
PROBLEM EXPLICATION IN THE PRINTING JOB When a sample order comes to the printing section usually it takes 2‐3 days to complete the job. The specific lithography machine requires at least 500 numbers of sheets to be printed until the best quality is acquired with a fresh colour refilling. Therefore, to check whether the colour is correct, the employee needs to produce batches of 500 tags. If the employee could not match the desired colour from the 1st batch, that batch will be wastage of money and time. When we consider the production rate of a machine is 3000 sheets per hour. It’s only a 10 minutes job which takes up to 3 days to complete. In decision making we identified this section as the bottle neck of the whole process. We focus mainly in printing section to reduce the lead time of the whole process. Main reason why sample printing process time consuming, 1. Usage of primitive colour recognition methods 2. Machine and labour availability 3. Absence of skill workers 4. Mechanical faults In order to prioritise the parameters, we use the method called “Pareto Analysis”. Pareto analysis is a statistical technique in decision making that is used for selection of a limited number of tasks that produce significant overall effect. It uses the Pareto principle: the idea that by doing 20% of work, 80% of the advantage of doing the entire job can be generated. Or in terms of quality improvement, a large majority of problems (80%) are produced by a few key causes (20%). According to their statistics the production rate is very much low than the expected value in the previous month. EXPECTED WORKING RATE (sheets per hour) 4 colour printer 2 colour printer die cutting machine guillotine cutting machine _______________________________________________________________________________________D OME Page 5 of 18 ME4042/09/01 3500 3000 4200 2200 ACTUAL WORKING RATE (sheets per hour) 1876 1747 2170 1400 MACHINE IDLE RATE 46.4% 41.76% 48.33% 36.36%
MACHINE
TIME STUDY After identifying the slow process we decided to perform a time study observe the time consumption for each step. STEPS TIME (seconds) Production Line 2 Production Line 3 Production Line 4
Selecting the colour coordinates From the colour card
‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ Mixing the colours Colour matching _______________________________________________________________________________________D OME Page 6 of 18 ME4042/09/01 Matching colours according 157 183 294 168 289 301 269 234
To the colour co‐ordinates of the Sample details given by the Design centre
SOLUTION FOR COLOUR MIXING AND COLOUR MATCHING PROCESS Colour mixing and colour matching process is the most essential requirement in the sample printing process as well as the mass production printing. Colour matching is a process which proper experience is a must and there is only one employee who is allocated for this sample making process. Once a sample unit is printed it is kept for a while to dry out the ink before colour matching. After that one of the best samples carries out for colour checking. As the data indicates this process takes up to 20‐ 30 minutes. In this time period the printing machines and the workers are jobless. Machine idling is a severe loss considered to production rate. Another major aspect of this situation is it varies the working rhythm of employees. This mentally affects the working efficiency of the employees. Since there is only single employee to perform colour checking as well as colour mixing, once a batch of 500 sheets is printed colour lab involves to the process for colour checking. Our solution is to improve the interaction between the colour lab and the printing section. In order to make the sample making process more sustainable and cog free, more trained personal should be recruited. Currently there is only one operator that is working as the colour co‐ordinator. This is a major disadvantage that the company is having as if the particular employee is not available, the whole process of sample making cannot be conducted. The main issue in recruiting a personal is the lack of experience. Currently the process is carry on mainly on experience of the colour co‐ordinator. Our solution to follow on the process without completely depending on experience of the employee, the process must be standardized. Coming up with a suitable colour data base is a way to make the colour matching process easy. There are few state of the art colour reading and mixing instruments costing around 200‐250 USD, that can be effectively used. When matching the colours the wastage can be minimized. A single colour barrel holds up to 4lb (1.81Kg) is 13.5USD In every sampling mix, approximately 400g of paint is used, that adds up to 14.91USD if four batches were used to get the correct combination of colours. By implementing a accurate colour adding methods nearly 70% of the sampling cost can be downsized. With standardizing the process it allows the company to recruit extra employee without focusing on the experience. By employing another hand the working time can be split in to half of him, resulting in 18‐24 hour shifts between them. Also any delays due to absence of the employee can be averted. _______________________________________________________________________________________D OME Page 7 of 18 ME4042/09/01
REDUCE TIME TAKING FOR INK DRYING The current method to dry ink is to keep aside printed materials in normal condition for 20‐25 minutes. We noticed this has another disadvantage of polluting the air around the working environment. Certain inks are harmful if you are exposed to inhale for a long time. Though ink does not easily cause death, inappropriate contact can cause effects such as severe headaches, skin irritation, or nervous system damage. Considering the health of the employing which directly affect to the growth of the company we introduced a proper ventilation system would be appropriate to create a friendly working environment. We recommended keeping the printed materials in an area which has a nearby window and maintaining the air flow with use of fans. MANAGING INK WASTE There are many practical ways for sheet fed and web lithographic printers to reduce the volume of waste ink generated: 1. 2. 3. 4. 5. 6. 7. 8. Ink Management Techniques for Better Reuse and Recycling Opportunities can be used to maximize the opportunities for ink reuse and recycling. Do not mix small quantities of leftover or obsolete inks with different colours of ink. Keep different types of ink separate. Store excess ink in properly sealed and labelled containers. Place plastic or waxed paper on top of sheet fed ink, and/or spray the ink with an anti‐skinning agent, or cover the ink with an oil consistent with printing inks to prevent oxidation. Do not dip knives deeply into sheet fed inks. Removing the ink evenly from the top surface of Transfer used ink back to the original empty containers and prevent drying by keeping the ink the ink can reduces the surface area of the ink exposed to oxidation. containers sealed. Help press operators to accurately estimate the amount of ink needed for each job through training in ink estimating techniques. Keep accurate records of the quantity of ink that is used for specific jobs, particularly for repeat customers' jobs or re‐orders. Use a standard ink sequence ‐ from light to dark ink. Monitor your ink inventory and use existing stock according to the "first in ‐ first out" strategy. Test any out‐of‐date ink for usability before you consider it waste ink. Carefully label, log, and store special‐order colours for future use rather than dumping them into waste ink drums. Donate ink that you no longer use to schools, or give the ink to other printers, rather than pay for disposal. (Colleges, universities and vocational/tech schools with graphic arts programs often have small on‐site print shops.) Use an automatic ink leveller to maintain the desired ink level in the fountain. Dedicate presses to specific colours or special inks to decrease the number of cleanings required for each press. Keep in communication with ink suppliers regarding proper use and handling procedures for their inks.
_______________________________________________________________________________________D OME Page 8 of 18 ME4042/09/01
Clearly mark the containers used to collect waste ink to prevent mistakenly discarding it. Avoid Don't treat excess ink as waste. Instead, manage it like a manufacturing by‐product that should be re‐introduced, as much as possible, back into the manufacturing system
contamination with solvents and trash (e.g., floor sweepings, cigarette butts, etc.).
WORKER FRIENDLY WORKING ENVIRONMENT There are altogether 13 printing and cutting machines. Each machine is operated by a single worker. The fact we noticed from our time study is time consuming for the setting up the machine. While operating the printing machine the employee has to move out printed materials. This change the working rhythm of the employees compared to repetition movements which is more efficient. We recommend recruiting a trainee as an assistant to machine set up. At the mean time he can be used to move printed materials. Other than working as individuals this indirectly affects to the boring nature of the job and creates a nice working environment. In the current process each day employees are working in different machine. For a instance there are 7 printing machines and workers are not allocated to a specific machine due to day shifts and night shifts of work. This reduces the proper maintenance of the machines and employees are not satisfying with machine repairs. We thought of allocating a group of people for a production line both for day and night shifts. For a example production line 1 has 2 printing machines and 2 cutting machines. Allocating 4 people for printing machines and another 4 for the cutting machine for both shifts are appropriate. This also improves the ability to work as a team and taking a responsibility. MACHINE AVAILABILITY Another major problem is the machine availability. The sample printing is carrying on parallel with the mass printing process. So once mass printing is occupied by a printing machine that machine is not available for entire day. Unlike a domestic printer once a ink refilling is performed it uses until the job completes. So finding a time slot for machine is a main issue. Managing time slots for machine allocation is done by 2 employees with the supervision of the production manager. The interaction between the management and the printing section is insufficient. Because of that the management is not updating frequently about the printing section. _______________________________________________________________________________________D OME Page 9 of 18 ME4042/09/01
PREDICTED RESULTS Cost benefit analysis for Inc recycling Average cost of a 4lb of lithography paint Average cost per ink mix for sampling Number of sample batches per process Reduction of Ink paint per sample Reusing and recycling amount = 200g = 30g = 230g (average) = 690g = 5.146 USD
. .
= 13.5USD = 400g = 5.2 USD = 3 to 4 = 15.6USD
Amount of paint/ ink used in a simple sample
Total cost per a sampling process of 3 ink mixes
Total amount of ink saved per sample mix For 3 samples Saved amount Saved amount as a percentage
=
×100% = 33%
Cost benefit analysis for a batch process
Data Finished samples per month Number of pieces per sample Sample avg. Cost per 1000 pcs Sample rejection rate Lead time for a single sampling job For the current sampling process Samples rejected per month Total losses due to rejection For a month Lead time for a single sampling job Number of samples finished per Number of parallel operations Number of successful samples Month without parallel operations = = = = = = 40 6000 12USD ( Rs. 1380.00 30% 12 days
_______________________________________________________________________________________D OME Page 10 of 18 ME4042/09/01
For the expected sampling process According to our objectives we are hoping to improve quality control of the process and reduce the rejection rate up to 20%. Then reduce the lead time up to 9 days. Expected sample rejection in the current process = 40 × 0.2 = 8 samples due to improved quality control Total losses due to rejection for a month Approximate reduction of losses = = = Rs. 8 × 6000 × Rs. 66240.00 Rs. 33120.00 = = = ≈ = ≈ = = 50 % 3.33 samples 16 × 3.33 53 samples 53 × 0.8 42 samples %
After reducing the lead time number of samples Finished per month without parallel operations Total number of finished samples per month with 16 parallel operations Number of successful samples
Improved rate of successful samples
_______________________________________________________________________________________D OME Page 11 of 18 ME4042/09/01
STAFF SCHEDULING AT THE LITHO PRINTING DEPARTMENT Another methodology of making the sampling process of the department of litho printing is staff scheduling/personal scheduling. Effective personnel scheduling has become one of the primary means by which service organizations remain competitive. Poor personal schedules can lead to an oversupply of workers with too much idle time, or an undersupply with an attendant loss of business. In this methodology, we present an integrated model that can be used to find optimal schedules for a homogeneous workforce. The objective is to meet daily staffing requirements at minimum cost without violating labour agreements. The approach requires the solution of a large‐scale integer linear program to determine general staffing needs for both full‐time and part‐time employees. Weekly tours are then constructed, and each employee is assigned a lunch break. Extensions to the model include different days of policies, variable start times, and the use of part‐time, flexible workers. The first component of the problem involves shift scheduling. The corresponding objective is to find the optimal crew size and daily work assignment for each member of the crew each day of the week. A shift is a set of consecutive time periods within a day and its length is the total amount of time it covers. In this study, a shift may vary in length depending on whether it is associated with a full‐time or part‐time employee, but cannot extend into the following day. This introduces the discontinuity property. The shift scheduling problem begins with the definition of permissible shifts, and concludes with the number of employees that should be assigned to each so as to satisfy daily demand period by period. The second part of a weekly schedule requires the specification of the days off. The number and characteristics of those days (weekend days, weekdays or combinations thereof) vary according to the organization and the type of industry in which it operates. A general consideration is that sufficient slack must be provided throughout the week so that the day off requirement is satisfied for every worker. The shift scheduling problem has to take this requirement into account when determining the optimal workforce size. An efficient way of doing this is by introducing lower bound constraints in the shift scheduling model. The third component of a bid job is the lunch break. All shifts, except those shorter than a specific number of periods, require a break. Labour contracts determine the start time and the duration of this break. General practice is to create a break window for every shift: a set of consecutive periods during which a break may be given and to assign a break within the window. Depending on the nature of the work being performed, the breaks could be staggered or coincident. Because an employee at lunch is off the clock, there should be sufficient resources to cover for him. Therefore, the schedule has to provide the necessary amount of over coverage, as well as make sure that everybody is taking his break within the prescribed window. The implicit modelling of break allowances for each employee is possible with the appropriate variables and constraints. However, this approach will only guarantee that there are a sufficient number of idle periods for every worker, i.e., the required slack is available. We used a circular 0–1 matrix for the general consecutive days off requirements associated with the (k,m) cyclic staffing problem in which employees are supposed to have k out of m days off. In the formulation, the circular matrix contains all possible patterns for k off‐days with a “1” denoting the day when the worker is on, and a “0” the day when the worker is off. The solution provides the optimal number of employees for each shift type together with their off‐day assignments. A number of solution techniques were considered including a linear non‐singular uni‐modular transformation of the decision variables, the conversion of the integer programming model into a series of network flow problems, and a linear relaxation followed by a round‐off algorithm. _______________________________________________________________________________________D OME Page 12 of 18 ME4042/09/01
MODEL DEVELOPMENT The baseline model for the shift scheduling problem was built in litho printing department. This facility has three types for workers as full‐time regulars (FTR), part‐time regulars (PTR) and part‐time flexible (PTF). A regular employee has a constant start time for every working day. A flexible (part time) employee is not given a 5‐day schedule but is called in when needed. He or she can have different start times on each day worked. Employees and managers generally prefer a constant start time for the entire week because of the consistency it offers. Because we wish to consider the simplest situation first, the entire workforce in the baseline model is assumed to be composed of regular employees only. PTFs will be included when various scenarios are examined. A full‐timer works 8 ½ consecutive hours which includes a 1/2 h allowance for a lunch break (in reality, he is off the clock for the ½ lunch) . A part‐timer, on the other hand, may be assigned one of a variety of possible shift lengths. In this study, we consider five different lengths from 4 to 8 ½ h (including the lunch breaks where applicable). All employees working 6 or more hours per day must be given a ½ h lunch break. The facility operates 24 h a day, 6 days a week (except Sunday) and is driven by service standards for processing the mail within a fixed time of its arrival. The schedules for regular workers are determined through a bidding system but the schedule for the flexible part‐timers can be changed weekly. In cases where the entire workforce is not sufficient to meet demand, temporary workers or overtime is used. The model discussed in this study is for the long‐range planning problem rather than the weekly scheduling problem. In the baseline model, the workday is divided into 48 periods, each 30 min long. The first period starts at 8:00 a.m. The department uses three intervals in a day for managerial purposes. Shifts are evenly distributed in these intervals. Shift type Full‐time Part‐time A regular employee starts during one of these three intervals and works a shift of a predefined length. The baseline model includes 2 different full‐time shifts whose start times are listed in Table 1. A full‐ time shift is 17 periods (8 ½ h) long and includes the lunch break. There are 3 different start times for a part‐ time shift and 2 different lengths for each start time, making 6 different part‐time shift types in all. Allowable part‐time shift lengths are 9 and 17 periods, including the breaks for 17 period shifts. The breaks are typically assigned sometime between the 9th and the 12th period giving a break window of 4 periods or 2 h in length. Also, every worker must be given 2 days off a week there are no restrictions in this regard but two consecutive days off, or at least one Saturday or Sunday, is preferable. The following notation is used in the developments. Start times 8.00 am / 7.00 p.m. 8.00 am /3.00p.m./ 7.00 p.m. Shift lengths 8 ½ h (17 periods) 8 ½ h, 4 ½ h
_______________________________________________________________________________________D OME Page 13 of 18 ME4042/09/01
Indices d t,s f p Parameters Cf Cp Gft Ppt Ddt K Q nF nP ρ Sets Bfs Bps Ffs Fps T M N Decision variables Xfd Ypd Βdt Wf Vp _______________________________________________________________________________________D OME Page 14 of 18 ME4042/09/01 number of employees assigned to full‐time shift type f on day d number of employees assigned to part‐time shift type p on day d total number of breaks initiated in period t on day d total number of full‐time employees needed for shift type f total number of part‐time employees needed for shift type p { j: break window for full‐time shift j lies entirely between period s and q } { j: break window for part‐time shift j lies entirely between period s and q } { j: break window for full‐time shift j lies entirely between period k and s } { j: break window for part‐time shift j lies entirely between period k and s } set of part‐time shift types that have breaks set of initial periods of the break windows, in ascending order set of final periods of the break windows, in ascending order prorated weekly cost of full‐time shift f prorated weekly cost of part‐time shift p 1 if full‐time shift type f covers period t; 0 otherwise 1 if part‐time shift type p covers period t; 0 otherwise demand for period t on day d earliest period a break can begin for any of the permissible shifts latest period a break can begin for any of the permissible shifts number of full‐time shifts number of part‐time shifts full‐time to part‐time labor ratio Index for the days of the week; d = 1; : : : ; 6 index for time periods during a day; t = 1; : : : ; 48 index for the full‐time shift types; f = 1; : : : ; nF index for the part‐time shift types; p = 1; : : : ; nP
Objective function (1a) minimizes the total weekly cost of the workforce. For full‐time shifts, the cost coefficients cf are all the same because the shift lengths are constant. On the other hand, part‐time shifts have a number of possible shift lengths, so cp has a different value for each part‐time shift p. If the cost parameters for any shift type differ according to the day (weekend days may cost more), then the variable definitions and objective function need to be modified slightly. The first constraint, (1b), assures that the net workforce is sufficient to cover the demand for each period, every day. The net workforce is the total number of part‐time and full‐time employees whose shift definitions cover that specific period (i.e., the workers who are responsible for that period of the day), less those who have a break during that period. The 0–1 matrices (G and P) filter out the shifts that do not cover the period under consideration. Constraint (1c), sometimes referred to as the ratio constraint, limits the number of part‐time employees. In general, part‐timers cost less than full‐timers because they have less seniority and may not get some benefits. If there are no limits on the number of part‐time employees, no full‐time employees would be chosen by the model. This is an important issue given that almost every union contract has an upper bound on part‐timers defined in various ways. This bound is determined by head counts on the payroll. A ratio based on the total hours worked instead of the number of employees might be an alternative. Constraints (1d)–(1g) are used to calculate lower bounds on the number of workers required to meet the daily demand. The first of these bounds, L1, is needed to assure that there is enough coverage so that every worker can _______________________________________________________________________________________D OME Page 15 of 18 ME4042/09/01
take 2 days off a week. Constraints (1d) and (1f) correspond to L1 for the full‐time and part‐time workforce, respectively. The second lower bound, L2, is necessary to assure that there are a sufficient number of workers to cover the day with the highest demand. Constraints (1e) and (1g) correspond to L2 for the full‐ time and part‐time workforce, respectively. Mathematically, the lower bounds can be represented as follows for each shift type f (similar constraints are applicable for shift type p):
1
1 5
2
:
1…6
Where Xfd denotes the number of workers assigned to shift type f on day d. To account for breaks, three more constraints are needed. The first, (1h), is referred to as the forward pass constraint. It assures that the total number of breaks initiated form period k up to a given period s exceeds the total number of employees who should have taken their breaks by that period. The employees included in the constraint are those whose break windows are fully covered through s, but not the ones who have the option of a break in some future period. The second constraint (1i) is referred to as the backward pass constraint and assures that the total number of breaks that are initiated from some specific period s through the end of the day (or until the last period that can be taken as a break, which is q) exceeds the number of employees who are entitled to a break during this interval. In other words, there should be sufficient breaks in the future to satisfy the break requirement for the rest of the day. These two constraints are needed to provide every employee with a break, but they are not sufficient to enforce the requirement that exactly one break be assigned to each worker entitled to one. Furthermore, they do not limit the break assignments to their respective ranges. Constraint (1j), which is known as the balance equation, is needed to assure that every worker is assigned a break and that it is within its permitted window.
Data No. of full time employees No. of part time employees = = = 344 27 Rs.4500/=
prorated weekly cost of full‐time prorated weekly cost of part‐time shift 4.5hr shift 8.5 hr
= = =
Rs.2250/= Rs.4500/= 344 27
full‐time to part‐time labor ratio
_______________________________________________________________________________________D OME Page 16 of 18 ME4042/09/01
SUMMARY As the project main objective was to reduce the process time of litho printing sample making process firstly we analyzed the problem thoroughly and we found following as the reasons to make sampling process time consuming 1. Usage of primitive colour recognition methods 2. Machine and labour availability 3. Absence of skill workers 4. Mechanical faults In order to prioritise the above parameters, we used the method called “Pareto Analysis” to determine the tasks that produce significant overall effect. And we performed a time study to analyse time periods for each operation and we have done a staff scheduling to meet the best time allocations for each job. Out of all above reasons the bottle neck was the colour matching process, the report includes the solutions we provided to make colour matching process less time consuming and accurate. Besides the material wastage (Ink) was concerned when using primitive colour recognition methods. This report includes all the analyses we have performed, all the data we obtained and solutions provided to achieve our objectives. _______________________________________________________________________________________D OME Page 17 of 18 ME4042/09/01
REFERENCES [1]. Aykin T. (1996) Optimal shift scheduling with multiple break windows. Management Science.p. 591–602. [2]. Pareto Analysis Step by Step URL: http://www.projectsmart.co.uk/pareto‐analysis‐step‐by‐step.html [3]. How to Perform a Time Study URL: http://www.howtodothings.com/business/how‐to‐perform‐a‐time‐study
_______________________________________________________________________________________D OME Page 18 of 18 ME4042/09/01
