4
Findings 2×P1 2×P2
3,814
WhatsApp msgs (window)
11/19
Machines with any output
42
Report rows flagged
2,128
Digital inward total (taka)
26
Defect mentions (window)
Digital grey inward
Empty slot = sheet not posted/received (OPT-001). 09-04 recovered via backfill — sheet exists despite FOSTTA holiday. Click to expand.
Watch list (active now)
- 09-19 "Chalu nhi ho rha" PC (unidentified) — fix reply pending; check next run.
- Ductable coil replacements (2/3) + Ricoh compressor arrival — dates pending.
- Mc1 09-18 stop: no further status found in window; presumed resolved or superseded — confirm closed.
- 09-18/09-19 production reports not yet posted at data end (intraday-normal).
Remark · per management (09-18), motherboard failures are in practice power-related; IT states them without cause ("Motherboard problem", "power line repaired"). Power-quality hypothesis leans true — pending voltage-logging before any stabilizer purchase.
Priority findings
P1 OPT-003HVAC — pre-planning + spares policy delivered 09-19 (5 days late); installs pendingOPEN
HVAC chronic: cooling failures → colour cut + machine stops; 09-19 pre-planning+spares policy delivered (2/3 ductable coils to replace, warranty-only compressors), design-room split Done 09-18; installs (coils/compressor) pending, response SLA unmeasured
Evidence
09-18 chase listed 4 demands (paper-print coil "very very very important", design-room split, Ricoh compressor, written preplanning). Same day: "Split A.C Design room Done running ok". 09-19 02:56 pre-planning text: 3 unit kinds (Carrier ductable / Mitsubishi+OGeneral split / OGeneral window); 2 of 3 ductable coils to be replaced; one spare condenser coil + one coil set each for Mitsubishi/OGeneral/window; compressors warranty-only; PCB/remote 2-day repair, no stock. So the EOD-09-14 schedule demand is now answered in writing; coil/compressor installation dates still unrecorded. Single-source (AC group text).
Loss Type
Availability + Quality chain (unchanged)
Priority
P1 (unchanged)
Longevity
chronic (unchanged)
Fix Class
structural (unchanged)
Owner
management (Aditya Khemka) + HVAC vendor
Next Check
coil + compressor installation dates posted; median complaint→onsite <24h; zero heat-attributable "colour cut"/"m/c band".
P1 OPT-004IT/PC — one same-day close (hang→Done 09-18); one new open ("Chalu nhi ho rha" 09-19); photos carry no serialsOPEN
IT/PC recurring hardware failures (UPS ×2 in 4 days, motherboard repeat failure); IT task open >1 month until escalation. Owner update 09-18: motherboard failures are power-related per IT's own reports ("power line repaired" phrasing) → power-quality protection (logging → SPD/servo + online UPS + earthing check) now part of this finding's fix path
Evidence
09-18 "Maru pc hang" → "Done" within ~15 min (close-out worked once). 09-17 paper-print m/c-1 monitor fault closed via "VGA slot not working, new HDMI connected, now working fine". New: 09-19 "Chalu nhi ho rha h" + photo, no fix reply by data end. 5 new device photos (09-18/09-19, both PC groups) show blank/black monitors + towers, HP/AOC/Logitech marks only — no serial, model label, or PC number legible, so **zero `pcsystem.db` inserts** (nothing to extract; text issues already covered by existing event pipeline, latest event 09-19 10:37). Single-source.
Loss Type
Availability (+ data) (unchanged)
Priority
P1 (unchanged)
Longevity
chronic (unchanged)
Fix Class
structural + habit layer (as run 1); add: sender must caption photos with PC number.
Owner
IT (Rehan + vendor)
Next Check
≤1 new complaint/week; UPS battery list with dates; IT report without chasing 2 weeks; photos captioned with PC number.
Machine utilization (window)
Bar = avg mtr/day on running days. Green ≥80% days active · amber ≥30% · red <30%. Click the chart to expand.
Findings
P1 OPT-003HVAC — pre-planning + spares policy delivered 09-19 (5 days late); installs pendingOPEN
HVAC chronic: cooling failures → colour cut + machine stops; 09-19 pre-planning+spares policy delivered (2/3 ductable coils to replace, warranty-only compressors), design-room split Done 09-18; installs (coils/compressor) pending, response SLA unmeasured
Evidence
09-18 chase listed 4 demands (paper-print coil "very very very important", design-room split, Ricoh compressor, written preplanning). Same day: "Split A.C Design room Done running ok". 09-19 02:56 pre-planning text: 3 unit kinds (Carrier ductable / Mitsubishi+OGeneral split / OGeneral window); 2 of 3 ductable coils to be replaced; one spare condenser coil + one coil set each for Mitsubishi/OGeneral/window; compressors warranty-only; PCB/remote 2-day repair, no stock. So the EOD-09-14 schedule demand is now answered in writing; coil/compressor installation dates still unrecorded. Single-source (AC group text).
Loss Type
Availability + Quality chain (unchanged)
Priority
P1 (unchanged)
Longevity
chronic (unchanged)
Fix Class
structural (unchanged)
Owner
management (Aditya Khemka) + HVAC vendor
Next Check
coil + compressor installation dates posted; median complaint→onsite <24h; zero heat-attributable "colour cut"/"m/c band".
Findings
P1 OPT-004IT/PC — one same-day close (hang→Done 09-18); one new open ("Chalu nhi ho rha" 09-19); photos carry no serialsOPEN
IT/PC recurring hardware failures (UPS ×2 in 4 days, motherboard repeat failure); IT task open >1 month until escalation. Owner update 09-18: motherboard failures are power-related per IT's own reports ("power line repaired" phrasing) → power-quality protection (logging → SPD/servo + online UPS + earthing check) now part of this finding's fix path
Evidence
09-18 "Maru pc hang" → "Done" within ~15 min (close-out worked once). 09-17 paper-print m/c-1 monitor fault closed via "VGA slot not working, new HDMI connected, now working fine". New: 09-19 "Chalu nhi ho rha h" + photo, no fix reply by data end. 5 new device photos (09-18/09-19, both PC groups) show blank/black monitors + towers, HP/AOC/Logitech marks only — no serial, model label, or PC number legible, so **zero `pcsystem.db` inserts** (nothing to extract; text issues already covered by existing event pipeline, latest event 09-19 10:37). Single-source.
Loss Type
Availability (+ data) (unchanged)
Priority
P1 (unchanged)
Longevity
chronic (unchanged)
Fix Class
structural + habit layer (as run 1); add: sender must caption photos with PC number.
Owner
IT (Rehan + vendor)
Next Check
≤1 new complaint/week; UPS battery list with dates; IT report without chasing 2 weeks; photos captioned with PC number.
Insights & suggestions
What the data shows
- UPS failures on 3 different PCs in 4 days (8/29, 9/14, 9/17) — one needed a battery change
- Repeat motherboard deaths (8/31 paper-print MC1, 9/7) — one back after 4–5 months with the same fault; management assesses power-related (IT states no cause)
- Procurement confirms the chain: 3 SMPS ordered 09-01 for poly-print PCs; motherboard chase same day; 09-17 RTC-section board repair quoted ₹1800+GST
- AC electrical signals: office-AC sparking (9/12), design-room compressor dead (9/16), Ricoh compressor undelivered, m/c stopped from heat (9/5)
- Cuts themselves are routine — the damage is what dies after them
PC fleet (IT register, Google Sheet 09-09)
44 PCs across 4 sections (+23 design-dev PCs in the 08-28 xlsx) · CPU generation mix:
| ≤2013 (gen ≤4) | 5 |
| 2015–18 (gen 5–8) | 12 |
| 2019–21 (gen 9–11) | 14 |
| 2022+ (gen 12+) | 10 |
| other/unknown | 3 |
Pre-2019 machines = 17 of 44 — aging boards + budget PSUs (ARTIS/ANT/EVM) tolerate dirty power badly. Register has no UPS column — add one, log PSU swaps per PC.
Power-quality decision plan (before buying anything)
1
Log the voltage — ₹0–8k, 2 weeks. Logger on main incoming + PC circuit; readings 9am / 2pm / 7pm / 11pm. Verdict: sustained 400V ±10% violation or deep sags → stabilizer justified; steady → skip it.
2
Failure forensics — ₹0. Compressor: burnt winding = electrical, dead capacitor = aging. UPS: swollen batteries = heat/age, fried SMPS = spike. Boards: dust + heat = thermal. PSU rail test (₹500 tester) on every failed PC before it leaves the bench.
3
Earth test — ₹2–5k. Bad earthing is the #1 hidden killer near stenter VFDs.
| If the log shows… | Then buy | Indicative cost |
|---|---|---|
| Unstable voltage (sag/swell) | Servo stabilizer — 150 kVA whole-plant, or 2×75–100 kVA segmented + 15–20 kVA PC room (segmented preferred; copper windings, ±1%, bypass panel) | ₹2.1–3.1 L · ₹2.8–4.2 L |
| Stable voltage, spikes after cuts | SPD surge protection at main panel only | ₹15–40k |
| PSU rails out of spec on failed PCs | Bulk PSU replacement policy: board dies → PSU replaced same ticket (priority: Vishl RTX3060, PRINTER PC, DIGI1/ABC23, JAGRUTI) | ₹2.5–4.5k / PC |
| UPS = dead batteries, boards = dust/heat | Bulk battery replacement + PC-room cooling + cleaning | < ₹25k |
| Bad earth | Earth pits + bonding | ₹10–20k |
Payback: last 30 days — 2 dead compressors, 3 UPS failures, repeat motherboard, 3 SMPS ordered (09-01), RTC-section board repair quoted (09-17). Preventing 2 failures/year covers the package in <12 months.
Defect mentions by type (window)
Keyword mentions across dagi + main groups. Parties hit most: Diwan ×3; Meesha, Mukesh, Saha, Kamal, Golobel ×2 each. Split across two groups = no single defect ledger (OPT-006). Click to expand.
White / finish lots
Total 2,266 taka (≈ 227 k mtr) across 129 lots, 08-28 → 09-06. Shortage complaints map here (OPT-009). Click to expand.
Findings
P2 OPT-006Defect stream — 2 new reports + 09-18 folding dugi on 4 lots; still zero closuresOPEN
Defect stream has no closure loop; whole-lot defects (Sep 10, Sep 15) with no disposition or mtr recorded
Evidence
09-18 dagi group "sunrise orgenja 44pana, 5–6 tako me yesa" (+2 photos); 09-19 main group "Korean checks fabric me heading out hai". White 09-18 folding report (p74 batch): SLS 2242 (5 dugi pieces), 2243 (12), 2244 (7), 2307-DB (5) — dugi-heavy batch across all 4 SLS/DB lots. No "fixed/closed" reply in any thread. Single-source.
Loss Type
Quality + data (unchanged)
Priority
P2 (unchanged)
Longevity
systemic (unchanged)
Fix Class
habit-dependent (unchanged)
Owner
QC/printing leads (Divyesh Patel, vinaymishra9)
Next Check
≥80% reports with closure reply + mtr affected; weekly rework meters.
Findings
P2 OPT-011Design loop — Zari/44-pana rounds 09-18 eveningOPEN
Design color-correction loop: 3+ rounds per design, chased till 11pm; files mostly unnamed
Evidence (Hybrid Design Group)
09-18 17:18 nine design images + "Zari"; 20:29–20:32 "left to right 44" ×3 + "narrow and finest possible… ensure enough Zari". Same spec vocabulary as run 1, late-evening chasing pattern repeated. Thin data (1 evening), single-source.
Loss Type
Performance (unchanged)
Priority
P2 (unchanged)
Longevity
chronic, provisional (unchanged)
Fix Class
habit-dependent (unchanged)
Owner
design lead + client coordinator
Next Check
correction rounds/design (baseline ~3); % standard filenames.
What would move the needle
- Pickup aging log (posted → picked, escalate >48h) — kills the "8 din se challan" class of problems (OPT-012)
- Color spec sheet per client (whiteness %, brightness reference) — cuts the 3-round correction loop (OPT-011)
- Lot closure rule: taka expected vs received before "complete" — settles shortages same day (OPT-009)
- Stop reason + restart time for every loop/stenter pause — already emerging via Altaf's posts; make it the norm (OPT-010)
Open-issues ledger (cross-run, all IDs)
| ID | First seen | Priority | Status | Summary |
|---|---|---|---|---|
| OPT-001 | 2026-09-18 | P2 | PARTIAL | Grey-inward stall — Digital scope 09-07→09-18 known (09-18 Digital 175 added 09-19; manifest seeded). Remainder: 09-12/09-13 never posted; 09-17 Digital −40 human verify; structural risk (manual lag + failed downloads hidden) open |
| OPT-002 | 2026-09-18 | P1 | OPEN | Production report loop: byte-identical repost (Sep 15→16 hole in day attribution, Homer base −3,018) + ~1 arithmetic slip per 4 days |
| OPT-003 | 2026-09-18 | P1 | OPEN | HVAC chronic: cooling failures → colour cut + machine stops; 09-19 pre-planning+spares policy delivered (2/3 ductable coils to replace, warranty-only compressors), design-room split Done 09-18; installs (coils/compressor) pending, response SLA unmeasured |
| OPT-004 | 2026-09-18 | P1 | OPEN | IT/PC recurring hardware failures (UPS ×2 in 4 days, motherboard repeat failure); IT task open >1 month until escalation. Owner update 09-18: motherboard failures are power-related per IT's own reports ("power line repaired" phrasing) → power-quality protection (logging → SPD/servo + online UPS + earthing check) now part of this finding's fix path |
| OPT-005 | 2026-09-18 | P2 | OPEN | Zero-run machines (Stenter 7/−6, Whinch M/C no 1, Fuzing 2) — idle vs unallocated unconfirmed |
| OPT-006 | 2026-09-18 | P2 | OPEN | Defect stream has no closure loop; whole-lot defects (Sep 10, Sep 15) with no disposition or mtr recorded |
| OPT-007 | 2026-09-18 | P2 | OPEN | Machine-name churn fragments series (Folding dispatch ×3 names; Ritesh (k.man) missing from CSV) |
| OPT-008 | 2026-09-18 | P1 | OPEN | IT procurement confirms PSU/board power chain: 3 SMPS ordered 09-01, motherboard chase, 09-17 RTC-section repair quote; 15/23 Digital PCs on budget PSUs |
| OPT-009 | 2026-09-18 | P1 | OPEN | White/finish lot shortages: 7 "taka kam" events in 3 weeks ≈ 700–1,000+ mtr unaccounted (1 taka ≈ 100 mtr), no reconciliation loop |
| OPT-010 | 2026-09-18 | P2 | OPEN | Processing stops: cotton-loop coil leakage cost a full day (09-06→09-07); recurring loop/stenter pauses |
| OPT-011 | 2026-09-18 | P2 | OPEN | Design color-correction loop: 3+ rounds per design, chased till 11pm; files mostly unnamed |
| OPT-012 | 2026-09-18 | P2 | OPEN | Dispatch grey pickups age >1 week ("party garam"); no pickup-aging view |
OPT-001/002/007 (data-hygiene, internal) are detailed here only; run report: data/insights/optimization/2026-09-19.md